| Time |
Nick |
Message |
| 04:54 |
pinesol_green |
Incoming from qatests: Test Success - http://testing.evergreen-ils.org/~live/test.html <http://testing.evergreen-ils.org/~live/test.html> |
| 07:39 |
|
TaraC joined #evergreen |
| 07:42 |
|
remingtron_ joined #evergreen |
| 08:13 |
|
akilsdonk joined #evergreen |
| 08:16 |
kmlussier |
Good morning #evergreen! |
| 08:17 |
remingtron |
Good morning kmlussier |
| 08:18 |
kmlussier |
remingtron: I worked on a 2.8 doc needs page last week. I'll throw the link out to the DIG list for tommorow. |
| 08:19 |
remingtron |
kmlussier: great, thanks! We have a DIG meeting today too, right? |
| 08:19 |
kmlussier |
I also should update the MassLNC vm so that it has the latest bug fixes. |
| 08:19 |
remingtron |
maybe folks will signup in prep for tomorrow |
| 08:19 |
kmlussier |
remingtron: Yes, I think so. |
| 08:20 |
remingtron |
I'll add the doc needs link to the agenda for today |
| 08:20 |
remingtron |
well, it's already on the agenda. cool |
| 08:21 |
kmlussier |
akilsdonk: I'm guessing some of these 2.8 doc needs are already being done by ESI? http://evergreen-ils.org/dokuwiki/doku.php?id=evergreen-docs:2.8_needs |
| 08:21 |
|
Newziky joined #evergreen |
| 08:23 |
|
ericar joined #evergreen |
| 08:25 |
kmlussier |
@weather 02771 |
| 08:25 |
pinesol_green |
kmlussier: The current temperature in Rumford, East Providence, Rhode Island is 31.1°F (8:25 AM EST on March 05, 2015). Conditions: Light Snow. Humidity: 94%. Dew Point: 30.2°F. Windchill: 32.0°F. Pressure: 30.03 in 1017 hPa (Rising). |
| 08:25 |
kmlussier |
Wow, it's snowing! That's new and different. |
| 08:28 |
|
mrpeters joined #evergreen |
| 08:32 |
csharp |
ok, so it looks like SMS notification events are being created every time a hold becomes "shelf ready" even if sms_notify is null - can others confirm that behavior? |
| 08:33 |
csharp |
we have about 2.5 million such events |
| 08:33 |
csharp |
select count(*) from action_trigger.event where event_def = 103 and target in (select id from action.hold_request where sms_notify is null); |
| 08:33 |
csharp |
^^ assumes that 103 is the id for SMS hold is ready notifications everywhere |
| 08:34 |
kmlussier |
csharp: Yes, I think I filed a bug on that ages ago. Let me check. |
| 08:34 |
csharp |
kmlussier: okay - good |
| 08:34 |
bshum |
There is a bug, yes. |
| 08:34 |
csharp |
been rubbing my eyes for the past half hour looking at the screen |
| 08:34 |
kmlussier |
No, my bug must have been for something else. But maybe bug 1096209? |
| 08:34 |
pinesol_green |
Launchpad bug 1096209 in Evergreen 2.4 "SMS functionality generates unnecessary event churn" (affected: 1, heat: 6) [Undecided,Triaged] https://launchpad.net/bugs/1096209 |
| 08:35 |
|
Dyrcona joined #evergreen |
| 08:35 |
csharp |
kmlussier++ # yep that's it |
| 08:35 |
bshum |
It's only been a known reported issue for two years, but whatever... |
| 08:35 |
csharp |
wow - I'm 2+ years late to the party! |
| 08:35 |
|
rjackson_isl joined #evergreen |
| 08:36 |
csharp |
we just enabled SMS with January's upgrade, so just now noticing |
| 08:37 |
|
mmorgan joined #evergreen |
| 08:37 |
bshum |
I never got to try the validator bits that tsbere mentions on that bug ticket. |
| 08:38 |
bshum |
The extra churn wasn't a big deal with other problems we were combating at the time I guess. |
| 08:38 |
kmlussier |
OK, the bug I was thinking of was bug 1016653 |
| 08:38 |
pinesol_green |
Launchpad bug 1016653 in Evergreen "System tries to send SMS holds notice when user has not selected this notification method" (affected: 1, heat: 6) [Undecided,Invalid] https://launchpad.net/bugs/1016653 |
| 08:38 |
|
julialima_ joined #evergreen |
| 08:39 |
eeevil |
csharp / kmlussier: all hook=hold.available defs are run when the event occurs. the code that says "this event happened" doesn't (can't) know about each def's validation criteria. |
| 08:39 |
bshum |
2.2 seems like forever ago. |
| 08:41 |
eeevil |
(just for context) |
| 08:41 |
csharp |
the issue that brought this up was that a library staff was trying to trigger an SMS notification for a patron who didn't set that at the time of hold placement |
| 08:41 |
csharp |
of course, I'm seeing that the event had already been created, so the staff member was confused as to why it wasn't working |
| 08:42 |
csharp |
(because in the past they've done that by adding email notification to an existing hold and checking it in) |
| 08:42 |
csharp |
eeevil: so tsbere's idea to use an existing validator won't help? |
| 08:43 |
* csharp |
assumes HoldNotifyCheck is the validator he was thinking of |
| 08:44 |
eeevil |
csharp: well, there are 2 issues here. 1) useless events that need a better validator 2) trying to make an event recur by checking in again |
| 08:44 |
csharp |
eeevil: right |
| 08:44 |
eeevil |
so, to the former, yes, there needs to be a better validator (if there isn't one). they're amazingly easy to write, btw |
| 08:44 |
csharp |
yeah, I'm seeing that |
| 08:45 |
eeevil |
for the latter, we'd have to have the checkin code change the shelf time and fire the hold.available events again for the hold. I don't believe either happens on "checkin of on-shelf hold" today |
| 08:46 |
csharp |
so basically, HoldNotifyCheck checks to see if either phone OR email OR sms notifications are there - looks like the task is to break them out into three separate checks |
| 08:46 |
eeevil |
and, that's also not necessarily desirable, because folks who did have the settings set will get dupes |
| 08:46 |
csharp |
eeevil: yeah - I'm not sure I want that |
| 08:47 |
eeevil |
you don't necessarily need to split it, just add a parameter of, say, notification_type, and check that in the validator (and if that's not there, do what it does today) |
| 08:48 |
eeevil |
or split it into 3. judgment call on the best route, really |
| 08:49 |
eeevil |
do we want to maintain one per notification type (add more as we add notification types) or maintain one with a bit more logic |
| 08:49 |
|
Shae joined #evergreen |
| 08:50 |
csharp |
eeevil: so where would notification_type come from? - it's not a field in action.hold_request |
| 08:51 |
csharp |
and in many if not most of our cases, phone_notify, email_notify and sms_notify will be not null |
| 08:52 |
Dyrcona |
csharp: It's figured out on the fly from the presence of phone number, text number/notification checkbox, and email notification. |
| 08:52 |
eeevil |
a/t event def parameters are just hard-coded values. they're ways of passing data into the validator/reactor/cleanup modules |
| 08:53 |
eeevil |
so on the sms def you'd set parameter: name=notification_type, value=sms |
| 08:53 |
csharp |
oh, I see |
| 08:53 |
eeevil |
and on the email def you'd set value=email |
| 08:53 |
csharp |
got it now |
| 08:53 |
csharp |
I wasn't moving back far enough |
| 08:53 |
eeevil |
then, in the validator, if that's set you'd check exactly the right field |
| 08:53 |
eeevil |
instead of saying "any will do" |
| 08:57 |
csharp |
looks like there's already a similar parameter on the email def |
| 08:58 |
csharp |
check_email_notify |
| 09:00 |
csharp |
so the HoldIsAvailable validator appears to already be doing what we'd want the HoldNotifyCheck to be doing |
| 09:00 |
eeevil |
that validator, btw, is terribly named |
| 09:00 |
csharp |
heh |
| 09:00 |
eeevil |
because it's doing a TON more than "is this hold on the shelf" |
| 09:01 |
csharp |
wait - maybe the problem is that there *isn't* a check_sms_notify parameter |
| 09:01 |
dbwells |
csharp: Yes, I think that is what tsbere was saying so long ago. |
| 09:01 |
csharp |
dbwells: that's even better ;-) |
| 09:02 |
* csharp |
adds 'name=check_sms_notify, value=1' to 103's event parameters |
| 09:02 |
csharp |
now to test |
| 09:08 |
csharp |
argh - it still created the event |
| 09:08 |
akilsdonk |
kmlussier: I'll update the 2.8 doc needs wiki with the features that ESI is writing docs for. |
| 09:08 |
kmlussier |
akilsdonk++ |
| 09:10 |
csharp |
oh, of course it created the event, because it's "any of the above" |
| 09:10 |
Dyrcona |
So opensrf.settings is C and not Perl? |
| 09:10 |
Dyrcona |
Looks like it is, but I want to make sure. |
| 09:11 |
Dyrcona |
I'm trying to figure out why it said we were at max child processes yesterday when I don't think we were. |
| 09:11 |
Dyrcona |
I have suspicions, but I'm looking through the prefork C code, and want to make sure I'm in the right place. |
| 09:14 |
|
maryj joined #evergreen |
| 09:14 |
gmcharlt |
Dyrcona: I'm pretty sure it's Perl |
| 09:15 |
gmcharlt |
upon checking, yeah, it is |
| 09:15 |
Dyrcona |
gmcharlt: Thanks, there's osrf_settings.c which I thought might have been more than just utility routines. |
| 09:16 |
eeevil |
csharp: it created the event because the validator doesn't run until the event is created |
| 09:17 |
eeevil |
all active hold.available events are created. then, they're validated by the validator |
| 09:18 |
eeevil |
but, now, it shouldn't (assuming that's all that was needed, and there are no bugs) actually attempt to send an sms |
| 09:19 |
|
kbutler joined #evergreen |
| 09:19 |
Dyrcona |
gmcharlt: Thanks for pointing that out. Saves me some pointless digging through C code. |
| 09:20 |
dbwells |
csharp: In short, the SMS event will exist, but it's state should be 'invalid'. |
| 09:20 |
dbwells |
If that flag does what it looks like it should do. |
| 09:20 |
Dyrcona |
This of course makes the situation make less sense. |
| 09:21 |
Dyrcona |
Oh, and it happened at midnight last night, too. |
| 09:24 |
jeff |
my laptop is cold enough that it hurts to rest my hand on it. huh. |
| 09:24 |
jeff |
@weather ktvc |
| 09:24 |
pinesol_green |
jeff: The current temperature in Traverse City, Michigan is -4.0°F (8:53 AM EST on March 05, 2015). Conditions: Partly Cloudy. Humidity: 91%. Dew Point: -5.8°F. Pressure: 30.44 in 1031 hPa (Rising). |
| 09:26 |
kmlussier |
Brrrr |
| 09:33 |
csharp |
eeevil: dbwells: the events are being created with "pending" status - is that expected? |
| 09:33 |
|
yboston joined #evergreen |
| 09:33 |
eeevil |
yes |
| 09:33 |
csharp |
ok, so when it processes, it will become invalid |
| 09:34 |
eeevil |
pending->collected->[valid|invalid]->reacting->reacted->cleaning->cleanedup->[complete|error] |
| 09:34 |
eeevil |
more or less... |
| 09:34 |
eeevil |
invalid is an end point |
| 09:36 |
* eeevil |
looks at the a_t.event ... pending->found->collecting->collected->validating->[valid|invalid+STOP]->reacting->reacted->cleaning->[complete|error] |
| 09:36 |
eeevil |
and 'error' can happen at any point as well, if there's an... error :) |
| 09:38 |
csharp |
eeevil: thanks for that summary - very helpful |
| 09:38 |
gmcharlt |
eeevil: unless the 'error' status gets applied in error |
| 09:38 |
gmcharlt |
... which case, things have really gotten messed up! ;) |
| 09:39 |
csharp |
@who made an error by assigning 'error' status to the error in error? |
| 09:39 |
pinesol_green |
TaraC made an error by assigning 'error' status to the error in error. |
| 09:39 |
csharp |
@praise [someone] |
| 09:39 |
* pinesol_green |
the upgrade came off brilliantly, and it's all because of maryj |
| 09:39 |
csharp |
maryj++ |
| 09:39 |
eeevil |
TaraC would never do such a thing, pinesol_green |
| 09:40 |
maryj |
:D |
| 09:41 |
Dyrcona |
jef: Go inside, fool. :) |
| 09:41 |
TaraC |
whaaaaaa? |
| 09:41 |
* Dyrcona |
can't type. |
| 09:41 |
* Dyrcona |
also has another, lovely, "it only happens to us," think to figure out. |
| 09:42 |
Dyrcona |
s/think/thing/ |
| 09:42 |
Dyrcona |
I'll just write my comments in code for the rest of the day. I make fewer typos that way. |
| 09:42 |
Dyrcona |
$me->has_a->problem->uniq("to us"); |
| 09:43 |
* mmorgan |
can't type either ... just tried to gerp a file. |
| 09:43 |
csharp |
mmorgan++ # ha! |
| 09:43 |
Dyrcona |
mmorgan: Try slurp instead. :) |
| 09:44 |
* csharp |
adds 'alias gerp="echo 'derpity derp derp'" to his .bashrc |
| 09:44 |
Dyrcona |
$I->can_say("it isn't the hold targeter.") |
| 09:44 |
csharp |
Dyrcona++ |
| 09:45 |
mmorgan |
csharp++ :) |
| 09:45 |
Dyrcona |
It didn't run since 4:15pm yesterday when we had to restart everything. |
| 09:45 |
Dyrcona |
I cleared the lock file at 8:30 or so this morning. |
| 09:46 |
Dyrcona |
I often typo "mvlc" as either "mvcl" or "mlvc". |
| 09:46 |
Dyrcona |
It can cause problems when messing with email accounts. |
| 09:46 |
* gmcharlt |
can top that! |
| 09:46 |
gmcharlt |
I can typoe EVERYTHING |
| 09:46 |
gmcharlt |
;) |
| 09:46 |
Dyrcona |
Of course. |
| 09:47 |
Dyrcona |
Actually, my personal favs are every for ever, or using the wrong there/their/they're, apparently. |
| 09:47 |
kmlussier |
mmorgan++ |
| 09:48 |
Dyrcona |
mmorgan++ # for gerping files |
| 09:49 |
Dyrcona |
"Me spill chuck woks grate. I needle gamma chicken." |
| 09:49 |
mmorgan |
Ha! karma for typos! I could really clean up! |
| 09:49 |
Dyrcona |
:) |
| 09:50 |
gmcharlt |
tangentially reminds of John Scalzi's late cat, Ghlaghghee (sound it out) |
| 09:51 |
Dyrcona |
bshum can ignore this bit. :) |
| 09:51 |
Dyrcona |
So, if I am to believe my logs.... |
| 09:52 |
Dyrcona |
At 00:15 last night, 40 settings drones were spawned in 5 seconds. And 25 of them had processed 300 requests each to be gone in about 55 seconds. |
| 09:53 |
Dyrcona |
At 00:15:02 my drone counter says 5 were running. |
| 09:53 |
jeff |
so what A/T granularity runs at 00:15? |
| 09:53 |
Dyrcona |
jeff: I'll look in a minute. Let me finish... ;) |
| 09:53 |
* jeff |
grins |
| 09:53 |
Dyrcona |
At 00:15:07 osrfsyslog, says "no children." |
| 09:54 |
Dyrcona |
At 00:16:01 my drone counter says 20 settings drones were running. |
| 09:54 |
|
graced joined #evergreen |
| 09:54 |
Dyrcona |
My maxchildren is 45 and that's what is in the osrfsyslog. |
| 09:55 |
Dyrcona |
This happened at 16:23 or so yesterday afternoon, too. |
| 09:57 |
* csharp |
's personal maxchildren is 2 |
| 09:58 |
* gmcharlt |
giggles |
| 09:58 |
gmcharlt |
Dyrcona: are you interested in spitballing? |
| 09:59 |
Dyrcona |
I might be, but I want to see what happened Saturday morning, first. |
| 09:59 |
gmcharlt |
sounds good |
| 09:59 |
Dyrcona |
I had to restart services around 10:40 last Saturday, too. |
| 09:59 |
* jeff |
waits patiently for the device which holds one of his TOTP generators to gain enough charge to power on |
| 10:00 |
Dyrcona |
jeff: Is your office heated? |
| 10:00 |
jeff |
Dyrcona: yes. |
| 10:01 |
Dyrcona |
Well, turn it up. :p |
| 10:03 |
|
Newziky joined #evergreen |
| 10:06 |
Dyrcona |
Well, whatever happened Saturday morning, I didn't run out of settings drones, though looks like I did at 00:15 on Sunday. |
| 10:07 |
jeff |
Dyrcona: http://youtu.be/nu0R96OZy6w?t=27s |
| 10:07 |
|
krvmga_ joined #evergreen |
| 10:08 |
krvmga_ |
after getting my EG 2.7.3 installation working yesterday (minus SpiderMonkey), i upgraded to 2.7.4 this morning following the upgrade instructions. |
| 10:09 |
krvmga_ |
i don't know whether it's enough to mention it here or if i should file an LP bug |
| 10:09 |
krvmga_ |
the make to compile EG 2.7.4 dumped me out in my EG 2.7.3 directory even though i started it in my EG 2.7.4 directory |
| 10:11 |
krvmga_ |
the upgrade instructions were for upgrading to 2.7.3 but it was simple enough to substitute 2.7.4 in the right spots. |
| 10:11 |
bshum |
"Sounds like you've got problems beyond your battery." |
| 10:11 |
krvmga_ |
but i wondered if something might have happened |
| 10:11 |
krvmga_ |
my settings-tester runs all okay but i'm getting the 500 error now |
| 10:12 |
bshum |
krvmga_: I would imagine that maybe you might have typed some "cd" .. path command wrong when you were trying it. The configure / make command should occur in wherever you started out. |
| 10:12 |
bshum |
So you must have combined it with something weird that landed you back in 2.7.3 |
| 10:13 |
Dyrcona |
yep, what bshum said. |
| 10:13 |
bshum |
In the docs, there's a line like "cd /home/opensrf/Evergreen-ILS-2.7.3" |
| 10:13 |
bshum |
So that might have sent you back to where you shouldn't have been |
| 10:13 |
bshum |
If it was me, I would re-run through the whole upgrade process again |
| 10:13 |
krvmga_ |
bshum: it's a possibility |
| 10:13 |
bshum |
And make sure things went properly |
| 10:13 |
krvmga_ |
and i think that's what i'm going to do :) |
| 10:14 |
bshum |
When I see 500 errors, I tend to start looking at my server logs to find out what's breaking. But I would start with a cleaner install before I get too far away from things. |
| 10:14 |
krvmga_ |
my father used to do photography work for a large amusement park near pittsburgh when i was young |
| 10:14 |
bshum |
And if spidermonkey is still broken, then I'd worry too about what else was broken. |
| 10:15 |
bshum |
Broken, but unseen, anyways. |
| 10:15 |
krvmga_ |
the park would give him books of complimentary ride tickets. |
| 10:15 |
krvmga_ |
he would give the tickets to me and my brother. |
| 10:15 |
Dyrcona |
I don't think circ even loads without spidermonkey. |
| 10:15 |
bshum |
Right. |
| 10:15 |
krvmga_ |
we'd go to the park when it wasn't busy, get on the roller coaster and never get off the ride; just keep handing the operator more tickets. |
| 10:16 |
krvmga_ |
i'm having this feeling of deja vu |
| 10:16 |
krvmga_ |
i'm getting back on the evergreen ride now. :) |
| 10:18 |
|
RoganH joined #evergreen |
| 10:26 |
krvmga_ |
http://paste.evergreen-ils.org/41?tidy=on&submit=Format+it%21 |
| 10:26 |
krvmga_ |
just for information: that paste is the error that shows up for me when running make as opensrf - re:SpiderMonkey |
| 10:27 |
kmlussier |
@coffee [someone] |
| 10:27 |
* pinesol_green |
brews and pours a cup of Kenya Nyeri AB Gichatha-ini Signature Series, and sends it sliding down the bar to RBecker |
| 10:28 |
Dyrcona |
krvmga_: You ran make -f src/extras/Makefile.install your-distro as root, right? |
| 10:28 |
krvmga_ |
Dyrcona: right |
| 10:29 |
Dyrcona |
And you did make -f Open-ILS/src/extras/Makefile.install your-distro for Evergreen? |
| 10:29 |
Dyrcona |
'cause that is where SpiderMonkey would get installed. |
| 10:29 |
jonadab |
Hmm... is there a greppable version of the docs available at http://docs.evergreen-ils.org/2.7/index.html like maybe in the repo somewhere? |
| 10:29 |
Dyrcona |
I usually sudo -i and install the prereqs for both OpenSRF and Evergreen before doing any of the other build steps. |
| 10:30 |
krvmga_ |
make -f Open-ILS/src/extras/Makefile.install debian-wheezy |
| 10:30 |
krvmga_ |
is the command i used |
| 10:30 |
krvmga_ |
as root |
| 10:30 |
Dyrcona |
Well, I'd do it again and look for errors. |
| 10:30 |
krvmga_ |
Dyrcona: will do |
| 10:30 |
kmlussier |
jonadab: The docs are available in the main Evergreen repo. |
| 10:30 |
Dyrcona |
Did you set up CPAN? And, did you choose sudo instead of local dir or whatever? |
| 10:30 |
jonadab |
I'm looking in docs/ |
| 10:31 |
Dyrcona |
Maybe SpiderMonkey is installed in the wrong place. |
| 10:31 |
* Dyrcona |
waits on a zgerp to finish. :) |
| 10:32 |
jonadab |
Oh, I see. |
| 10:32 |
jonadab |
It's single-large-file, but it's there. And that may actually be better anyway. |
| 10:33 |
Dyrcona |
It finished, and that's when I notice the typo in the pattern..... |
| 10:33 |
* Dyrcona |
waits some more. |
| 10:35 |
Dyrcona |
It's typo kind of day! |
| 10:38 |
krvmga_ |
http://paste.evergreen-ils.org/42?tidy=on&submit=Format+it%21 |
| 10:38 |
krvmga_ |
Dyrcona: that paste is a fail i saw in the output |
| 10:39 |
Dyrcona |
And it isn't really a failure: "make_test FAILED but failure ignored because 'force' in effect" |
| 10:39 |
Dyrcona |
That gets installed anyway. |
| 10:40 |
Dyrcona |
krvmga_: Try this: find /usr/local -name SpiderMonkey.pm |
| 10:42 |
krvmga_ |
it finds the file in 3 places OpenILS/Utils (2 versions of perl) 5.10.1 and 1.14.2) and in JavaScript under lib /perl/5.10.1 |
| 10:43 |
krvmga_ |
1.14.2 -> 5.14.2 (typo) |
| 10:43 |
bshum |
eeevil: Speaking of fixed fields, is that related to the stuff you and gmcharlt were trying for in https://bugs.launchpad.net/evergreen/+bug/1306258 ? |
| 10:43 |
pinesol_green |
Launchpad bug 1306258 in Evergreen "more seed data for MARC21 fixed field values would be nice" (affected: 1, heat: 6) [Wishlist,Triaged] |
| 10:44 |
bshum |
Which I see now I forgot about during my 2.7 reign :( |
| 10:44 |
eeevil |
bshum: not in the case of Time, of course |
| 10:44 |
gmcharlt |
yeah, most likely separate |
| 10:45 |
gmcharlt |
and yeah - the seed data continues to need to be updated; moar newer code to ease that will fall out of the tag table subproject |
| 10:45 |
eeevil |
the list post is about how a higher level of config is lacking. specifically, the tag+pos+length info for some fixed fields. time and tech are just the two mentioned |
| 10:46 |
kmlussier |
Can that be treated as a bug fix? It would be nice to see it in 2.8, at a minimum. |
| 10:46 |
eeevil |
kmlussier: it's seed data, so, I suppose it's berick's call. it's relatively straight forward, but time consuming |
| 10:47 |
Dyrcona |
krvmga_: I don't know exactly where it should be on wheezy, but that sounds about right. |
| 10:48 |
Dyrcona |
Looks like it is installed. |
| 10:48 |
bshum |
Oh I remember this now. I got distracted with trying to built up an upgrade script for the branch and then... otherwise, distracted :) |
| 10:48 |
eeevil |
one would need to comb through http://www.loc.gov/marc/bibliographic/bd008.html http://www.loc.gov/marc/bibliographic/bd006.html and http://www.loc.gov/marc/bibliographic/bdleader.html looking for unrepresented fields to insert into config.marc21_ff_pos_map |
| 10:48 |
krvmga_ |
Dyrcona: yes, it seems right. |
| 10:48 |
krvmga_ |
so the "Please install..." message just leaves me scratching my head |
| 10:50 |
eeevil |
krvmga_: is "lib /perl/5.10.1" a typo? I'd expect to see it in 5.14.2 along side the OpenILS modules |
| 10:50 |
bshum |
"upgraded" systems can be so much fun. |
| 10:50 |
bshum |
By fun, I mean, BLOW IT UP and start over :D |
| 10:50 |
bshum |
Kidding, kidding... |
| 10:50 |
bshum |
(or am I?) |
| 10:51 |
krvmga_ |
eeevil: i actually see two directories there /usr/local/share/perl/5.10.1 and /usr/local/share/perl/5.14.2 |
| 10:51 |
jonadab |
bshum: No reason to blow anything up. fdisk will do the job nicely :-) |
| 10:51 |
krvmga_ |
both of them have OpenILS/Utils/SpiderMonkey.pm under them |
| 10:52 |
krvmga_ |
the 3rd directory is /usr/local/lib/perl/5.10.1/JavaScript/SpiderMonkey.pm |
| 10:52 |
eeevil |
krvmga_: yes, but I would expect to see JS/SM.pm under 5.14.2 if you ran the prereq installer after perl was upgraded (which it has been -- that's why there are 2 dirs) |
| 10:52 |
eeevil |
IOW, I suspect that perl does not see anything in the 5.10.1 dir |
| 10:52 |
jonadab |
(But yes, after about three consecutive upgrades of almost anything, I usually prefer to do a fresh clean install at some point and copy/migrate data in.) |
| 10:52 |
Dyrcona |
Perl should look first in the directory for its version and then go backward, but you can check with a one-liner: perl -e 'print join(":" @INC) . "\n";' |
| 10:53 |
Dyrcona |
And, mine only looks in 5.14 and 5.14.2 |
| 10:54 |
Dyrcona |
Missing comma in what I typed. |
| 10:54 |
Dyrcona |
As I said earlier..... ;) |
| 10:54 |
eeevil |
and you can test empirically for that module with: perl -MJavaScript::SpiderMonkey -e 'print $JavaScript::SpiderMonkey::VERSION."\n";' |
| 10:54 |
Dyrcona |
or just the -MJavaScript::SpiderMonkey if it doesn't load, Perl reports an error. |
| 10:55 |
krvmga_ |
eeevil: when i ran that command, i just got a ">" prompt in response |
| 10:55 |
Dyrcona |
Some modules don't havea VERSION variable. |
| 10:55 |
Dyrcona |
krvmga_: That means it is there and lacks a VERSION. |
| 10:55 |
Dyrcona |
It printed nothing. |
| 10:56 |
krvmga_ |
Dyrcona: it looked like it was waiting for more input. i had to Ctrl-C out of it. |
| 10:56 |
eeevil |
no, it means you missed the trailing quote |
| 10:56 |
Dyrcona |
Oh, yeah. Quotes.... |
| 10:56 |
* Dyrcona |
is not having the best of mornings. |
| 10:56 |
krvmga_ |
what eeevil said |
| 10:57 |
Dyrcona |
So, apparently, opensr.settings has been a problem for us for quite a while, at least all of this year. |
| 10:57 |
krvmga_ |
when i made sure the ' was at the end, i got Can't locate JavaScript/SpiderMonkey.pm in @INC |
| 10:58 |
krvmga_ |
(@INC contains: /etc/perl /usr/local/lib/perl/5.14.2 /usr/local/share/perl/5.14.2 /usr/lib/perl5 /usr/share/perl5 /usr/lib/perl/5.14 /usr/share/perl/5.14 /usr/local/lib/site_perl .). |
| 10:58 |
* Dyrcona |
quits trying to type and goes back to looking at log files. |
| 10:59 |
eeevil |
krvmga_: and you've run the prereq installers for both opensrf and evergreen during /this/ attempt at installation, correct? |
| 10:59 |
krvmga_ |
eeevil: i can do it all again. |
| 10:59 |
krvmga_ |
i'm getting good at doing everything over and over. |
| 11:00 |
eeevil |
just that one thing |
| 11:00 |
krvmga_ |
:) |
| 11:00 |
CheshireCat |
It doesn't particularly matter which way you go, if you don't know where you are going. |
| 11:01 |
krvmga_ |
looking at the output, it seems to be looking for SpiderMonkey in JavaScript/ rather than OpenILS/Utils |
| 11:01 |
krvmga_ |
am i reading that wrong? |
| 11:02 |
eeevil |
you are correct |
| 11:03 |
eeevil |
there is a module called JavaScript::SpiderMonkey. it wraps libjs (the "spider monkey" js engine from mozilla) |
| 11:03 |
eeevil |
we (evergreen) have a module called OpenILS::Util::SpiderMonkey that uses JS::SM |
| 11:03 |
|
vlewis joined #evergreen |
| 11:05 |
eeevil |
O::U::SM is included at the top level of several opensrf apps. circ is one. the tpac asks the circ app for ... stuff. but, the circ app never started because O::U::SM couldn't fine JS::SM and blew up |
| 11:05 |
eeevil |
so, you need to install JS::SM, which the prereq installer does for you, in a place that perl can find it |
| 11:06 |
eeevil |
krvmga_: just a wild guess: this used to be a squeeze box and you upgraded it to wheezy, yes? |
| 11:06 |
* Dyrcona |
giggles at "squeeze box." Decide to play The Who when his Rolling Stones play list runs out. |
| 11:08 |
bshum |
Doesn't seem so wild of a guess |
| 11:09 |
eeevil |
bshum: no, it doesn't ;) (based on https://packages.debian.org/squeeze/perl and https://packages.debian.org/wheezy/perl) |
| 11:10 |
krvmga_ |
eeevil: yes, this used to be a squeeze box. |
| 11:11 |
krvmga_ |
but i upgraded to wheezy. |
| 11:11 |
krvmga_ |
i sleep better at night. |
| 11:11 |
eeevil |
hey, no complaints there. wheezy++ |
| 11:13 |
Dyrcona |
My prod. server was upgraded from Lucid to Precise, so has multiple Perl directories, too. |
| 11:14 |
eeevil |
Dyrcona: and JS::SM is in all, right? |
| 11:14 |
Dyrcona |
Yep. |
| 11:14 |
eeevil |
via prereq installer |
| 11:15 |
Dyrcona |
It shows up in 5.10.1 and 5.14.whatever, so gets found. |
| 11:15 |
krvmga_ |
i'm just gonna go for a do-over on the upgrade. |
| 11:15 |
krvmga_ |
i'll be back later :) |
| 11:15 |
Dyrcona |
krvmga_: If you can just do a fresh install of.... |
| 11:15 |
Dyrcona |
Oh well. |
| 11:19 |
|
dreuther joined #evergreen |
| 11:21 |
jonadab |
Hmm... When I am not logged in, the client shows "Not yet configured for the specified server." under workstation. When I click the Login button, that disappears, and is replaced with... absolutely nothing. Status still says "Testing hostname...", if that is relevant. |
| 11:22 |
jonadab |
(The srfsh login test works on the server.) |
| 11:22 |
jonadab |
Is there a log I can look at to figure out what is happening, or not happening? |
| 11:22 |
eeevil |
jonadab: using a self-signed cert? |
| 11:23 |
jonadab |
eeevil: Yes. I added an SSL Exception. |
| 11:23 |
eeevil |
ah, k |
| 11:23 |
eeevil |
easy stuff: check |
| 11:25 |
* jonadab |
wonders if he used the wrong thing in the rel= thing or whatever it was when building the client -- building a version obtained via git. |
| 11:25 |
jonadab |
But the client is built per the instructions, make rigrelease rebuild blah blah blah... |
| 11:26 |
jonadab |
So that should pick up the previously-specified value and use it again, shouldn't it? |
| 11:26 |
Dyrcona |
jonadab: Does your firewall allow you to talk to the server? Are the server processes configured and running properly? |
| 11:27 |
Dyrcona |
We use custom client versions built from git in production, training, and development. |
| 11:27 |
jonadab |
Well, I can talk to the web server on port 80... hang on... yes, on 443 as well, once I tell my browser the cert is ok. |
| 11:28 |
jonadab |
To be clear: Seamonkey, on the same computer as the staff client, can talk to the Apache on the server. |
| 11:28 |
jonadab |
And get back a test web page. |
| 11:28 |
Dyrcona |
Ah, that's your problem. |
| 11:28 |
Dyrcona |
You should get the Evergreen OPAC. |
| 11:29 |
jonadab |
What's the URL for that? |
| 11:29 |
Dyrcona |
Doesn't sound like Apache is configured properly. |
| 11:29 |
jonadab |
(The test webpage is one I installed, /openils/var/web/test.html |
| 11:29 |
jonadab |
) |
| 11:30 |
Dyrcona |
If you hit the root of your Evergreen with a web browser, you should get the OPAC. |
| 11:30 |
jonadab |
(Which I'm getting via https://evergreen/test.html) |
| 11:30 |
jonadab |
Oh, the root? |
| 11:30 |
jonadab |
No, that doesn't work. |
| 11:30 |
jonadab |
I mean, it does, but. |
| 11:30 |
jonadab |
There's no index.html there, or anything. |
| 11:31 |
Dyrcona |
https://training.mvlcstaff.org/ |
| 11:31 |
Dyrcona |
You get redirected. |
| 11:31 |
jonadab |
Oh, I see. Interesting. |
| 11:31 |
jonadab |
Well /eg/opac/home do anything useful here? [tests] No, no it does not. |
| 11:32 |
jonadab |
Hmm... there is a /opac/, but no index file there either. |
| 11:32 |
jonadab |
Maybe the catalog stuff did not get correctly installed? |
| 11:33 |
jonadab |
/openils/var/web/opac/ contains directories: common extras images locale skin theme |
| 11:33 |
bshum |
That's an older path. |
| 11:33 |
bshum |
For another catalog (that's no longer used) |
| 11:33 |
jonadab |
Oh. |
| 11:33 |
jonadab |
Ok, then. |
| 11:33 |
bshum |
The current catalog comes from /openils/var/templates/opac/.. |
| 11:34 |
bshum |
The staff client talks to /openils/var/web/xul/server/.. (which is a symlink of server to rel_x_x/server) |
| 11:35 |
jonadab |
Oh, if I go to https://evergreen/xul/server, that works. |
| 11:35 |
jonadab |
"Welcome to Evergreen", etc. |
| 11:35 |
jonadab |
Looks like a PAC, indeed. |
| 11:36 |
Dyrcona |
Ah... What bshum said.... |
| 11:37 |
bshum |
jonadab: Eh, it might look like one, but it really isn't. And it won't operate quite as you'd expect in a browser. |
| 11:37 |
jonadab |
Yeah, it doesn't. I noticed. |
| 11:37 |
jonadab |
Clicking on things, for example, does not do much. |
| 11:37 |
bshum |
Since it references XUL specific things that the staff client uses |
| 11:37 |
jonadab |
I see. |
| 11:37 |
bshum |
i.e. opening stuff in tabs in the client. |
| 11:38 |
Dyrcona |
jonadab: on the server do ls -l /openils/var/web/xul |
| 11:38 |
Dyrcona |
But, I think that is correct if /xul/server is working. |
| 11:38 |
Dyrcona |
I think your Apache config is not quite all there. |
| 11:38 |
jonadab |
server is a symlink to rel_2_7_3/server |
| 11:39 |
bshum |
Unrelated, I just added a new docs commit for b8515a74944bad8634e841b1961f6224a4afc662 to update the rel_2_7 docs to point at 2.7.4 for the upgrade notes. |
| 11:39 |
pinesol_green |
[evergreen|Ben Shum] Docs: Change references to release 2.7.4 - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=b8515a7> |
| 11:39 |
jonadab |
There's also 0msater_9238334. |
| 11:39 |
bshum |
I didn't port it to master, because it occurred to me that we probably don't care about it as much there. |
| 11:39 |
jonadab |
Dyrcona: Ok, I will look over the Apache config again. |
| 11:39 |
Dyrcona |
jonadab: What does your client think it's version is? |
| 11:39 |
Dyrcona |
I forget where to look for that, prefs.js or something? |
| 11:40 |
jonadab |
Dyrcona: About says rel_2_7_3 |
| 11:40 |
jonadab |
Help->About |
| 11:40 |
Dyrcona |
Well, then, that should work. |
| 11:41 |
bshum |
If there's no automatic redirect with apache, then maybe the default is still being applied instead of the eg.conf config? |
| 11:41 |
jonadab |
bshum: No, the only enabled "site" is eg.conf |
| 11:41 |
jonadab |
But maybe I edited it wrong, or something. |
| 11:41 |
bshum |
Gotcha |
| 11:50 |
jonadab |
Oh, this may be relevant: access.log only gets entries when using links on the server; when accessing anything from elsewhere, it goes into other_vhosts_access.log |
| 11:51 |
* jonadab |
stares at ServerAlias |
| 11:54 |
|
collum joined #evergreen |
| 11:55 |
|
jwoodard joined #evergreen |
| 11:57 |
|
sandbergja joined #evergreen |
| 12:07 |
Dyrcona |
jonadab: A lot of the new default config throws logs at syslog, 'cause many sites are now running a central logging server. |
| 12:10 |
jonadab |
Ah, yes, I'm seeing that now. |
| 12:12 |
|
jihpringle joined #evergreen |
| 12:12 |
|
bmills joined #evergreen |
| 12:14 |
yboston |
I had a tiny code change accepted into 2.8, is it too late to push the commit notes? (I already have access to push to the docs/ on the main repo) |
| 12:15 |
yboston |
also, normally by when should I have pushed my release notes? |
| 12:18 |
Dyrcona |
Release notes, if written by the code author, are usually put in the branch when it is pushed to working before pullrequest is added, but it does vary. |
| 12:18 |
Dyrcona |
Since 2.8 final is not out, and we don't do release notes for minor upgrades, I think you're safe. :) |
| 12:21 |
kmlussier |
yboston: It looks like the 2_8.txt release notes file hasn't been built yet, so you should be safe putting them in the RELEASE_NOTES_NEXT directory. |
| 12:22 |
kmlussier |
There are a couple of other features I found that still don't have release notes. I was planning to add them at some point. |
| 12:22 |
yboston |
thanks Dyrcona & kmlussier |
| 12:22 |
kmlussier |
yboston: Typically, though, you want to create them before the code is merged. :) |
| 12:23 |
yboston |
kmlussier: learned my lesson |
| 12:32 |
pinesol_green |
[evergreen|Yamil Suarez] LP#1403967: show 'subject heading thesaurus' value release notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=32e6401> |
| 12:32 |
|
buzzy joined #evergreen |
| 12:51 |
bshum |
Huh |
| 12:51 |
bshum |
That bug doesn't have a milestone. |
| 12:52 |
Dyrcona |
And "Squeeze Box" comes around on the play list.... |
| 12:52 |
* bshum |
is bothered by that |
| 12:52 |
* Dyrcona |
rocks out. |
| 12:52 |
Dyrcona |
heh. |
| 12:52 |
Dyrcona |
Add one. |
| 12:52 |
* bshum |
also decides to fix up LP milestones for next round |
| 12:55 |
Dyrcona |
So, it looks like when settings drones are hit, they get hit hard. |
| 12:56 |
Dyrcona |
They don't seem to stick around long, so they get 300 requests in rather short order. |
| 12:57 |
bshum |
berick: I added a new 2.8.0 milestone; wasn't sure if the RC would be its own milestone or if bug fixes towards the RC would also count towards the final 2.8.0. |
| 13:00 |
|
akilsdonk joined #evergreen |
| 13:01 |
berick |
bshum: that's perfect, thanks |
| 13:02 |
Dyrcona |
My rough calculation is we got at least 7,500 settings requests between 00:15:02 and 00:16:01 last night. |
| 13:02 |
Dyrcona |
If the logs and settings are to be believed. |
| 13:03 |
berick |
oh my. you'll definitely get a lot if you restart services, but that seems high |
| 13:04 |
Dyrcona |
There's an a/t running that kicks off every 15 minutes, but how many triggers would it take? |
| 13:14 |
berick |
on the server side, perl loads and caches the settings at listener start time, so i don't think spawning new Perl children would even cause a settings server hit. |
| 13:14 |
berick |
apache children would, though |
| 13:15 |
berick |
C loads and caches first thing, too |
| 13:16 |
Dyrcona |
Unfortunately, I don't know how many Apaches were spawned at the time, nor if anyone was making remote settings calls. |
| 13:16 |
berick |
you can't make remote settings calls |
| 13:16 |
berick |
you have to be on the private jabber domain |
| 13:19 |
berick |
hmm, we're only logging the API requestor info at DEBUG |
| 13:21 |
bshum |
Bleh |
| 13:21 |
bshum |
https://bugs.launchpad.net/evergreen/+bug/1428780 |
| 13:21 |
pinesol_green |
Launchpad bug 1428780 in Evergreen "Documentation 404" (affected: 1, heat: 6) [Undecided,New] |
| 13:22 |
* bshum |
doesn't know enough about the epub generation to know what to check |
| 13:23 |
gmcharlt |
bshum: I"ve done other poking of people as well; I don't think you need worry about it |
| 13:24 |
bshum |
gmcharlt: Okie dokey, yep, all caught up with twitter now :) Thanks! |
| 13:29 |
bshum |
csharp: There's a facebook post on the Evergreen ILS page timeline from someone who's trying to register for the mailing list (I think) |
| 13:29 |
* dbs |
knows about epub generation but can't do anything about the docs.evergreen-ils.org hosting |
| 13:29 |
bshum |
https://www.facebook.com/EvergreenILS/posts/425667347602733 |
| 13:29 |
bshum |
Not sure if there's anything weird with mailman going on |
| 13:30 |
bshum |
I'm deciding how to tackle the question itself though. "General support" sounds odd to me. |
| 13:32 |
gmcharlt |
bshum: yeah, one thing if they had inadverantly dropped the word "mailing list" |
| 13:33 |
gmcharlt |
another thing if they have ... expectations about what being subscribed to open-ils-general means |
| 13:33 |
csharp |
bshum: nothing in the moderation queue |
| 13:33 |
bshum |
gmcharlt: Yeah that's the part I'm a little nervous about. |
| 13:34 |
csharp |
@who uses a walled garden to support their open source ILS? |
| 13:34 |
pinesol_green |
jcamins uses a walled garden to support their open source ILS. |
| 13:34 |
gmcharlt |
bshum: my suggestion, if you the list admin token, is to ask them to PM you their email address and subsribe them manually |
| 13:34 |
gmcharlt |
and just leave it at that |
| 13:34 |
csharp |
in other news, I'm deflated a bit to see that we're still very slowly adding missed things from past upgrade scripts: https://bugs.launchpad.net/evergreen/+bug/1096209/comments/4 |
| 13:34 |
bshum |
gmcharlt: The what token? |
| 13:34 |
pinesol_green |
Launchpad bug 1096209 in Evergreen 2.4 "SMS functionality generates unnecessary event churn" (affected: 1, heat: 6) [Undecided,Triaged] |
| 13:35 |
Dyrcona |
Best place to get "support" is IRC, and you get what you "pay" for. :) |
| 13:35 |
gmcharlt |
bshum: mailman admin credentials for the open-ils-general mailing list |
| 13:35 |
csharp |
gmcharlt: I'm your huckleberry (or phasefx is) ;-) |
| 13:35 |
gmcharlt |
if not, I know a csharp you can ask :) |
| 13:35 |
Dyrcona |
csharp: run master. upgrades are a cinch. :) |
| 13:36 |
gmcharlt |
*snap* |
| 13:36 |
bshum |
gmcharlt: Ah, okay. Yeah I don't have that. So someone else like csharp would have to do that. |
| 13:36 |
csharp |
Dyrcona: it would've been missed in the numbered upgrade scripts too |
| 13:36 |
bshum |
Well it does say "not covered" I guess. |
| 13:36 |
bshum |
:) |
| 13:36 |
bshum |
Good find csharp ;) |
| 13:36 |
gmcharlt |
csharp: willing to spread it to a couple more webteam members? |
| 13:37 |
Dyrcona |
Oh, one of those.... |
| 13:37 |
csharp |
RAISE NOTICE 'Yo, sys admin - you need this right here!'; |
| 13:39 |
csharp |
gmcharlt: absolutely |
| 13:39 |
gmcharlt |
csharp: I'd like one then :) bshum too, if he wants it |
| 13:39 |
* bshum |
doesn't want it. |
| 13:40 |
bshum |
At least, not today at this time. :) |
| 13:40 |
gmcharlt |
:) |
| 13:41 |
csharp |
gmcharlt++ # sharing the load |
| 13:45 |
gmcharlt |
bshum: I've responded to that query on FB |
| 13:45 |
Dyrcona |
csharp: We (i.e. tsbere) probably set that up manually for us. |
| 13:45 |
bshum |
gmcharlt++ |
| 13:50 |
|
bmills joined #evergreen |
| 13:51 |
kmlussier |
csharp: I think/hope we've come a long way with upgrade notes since the 2.2 days when SMS was released. Do you find that we're still having issues where potential gotchas aren't being addressed in upgrade notes? |
| 13:51 |
* kmlussier |
also wonders if something should be added to http://docs.evergreen-ils.org/2.3/_sms_text_messaging.html so that sites are aware of it when configuring SMS |
| 13:53 |
csharp |
kmlussier: I'm just kvetching to emphasize the need for internal documentation. Not trying to indict the past lack of end-user docs or call any individual out. |
| 13:53 |
|
bmills joined #evergreen |
| 13:54 |
csharp |
kmlussier: and this wouldn't probably belong in end-user facing docs, just in the output of an upgrade script as a hint that something else needs doing |
| 13:54 |
csharp |
(IMHO) |
| 13:55 |
Dyrcona |
Well, we generally do that now. It wasn't standard practice before. |
| 13:55 |
kmlussier |
csharp: Yeah, I didn't think you were trying to indict anyone. I'm just trying to see what we should be doing to make the process easier for people. |
| 13:56 |
yboston |
heads up the DIG monthly meeting starts at 2 PM EST |
| 13:57 |
csharp |
kmlussier: just making sure things like that get postgresql RAISE NOTICE statements is enough for me ;-) |
| 14:00 |
|
bmills joined #evergreen |
| 14:00 |
yboston |
#startmeeting 2015-03-05 - DIG Monthly Meeting Evergreen Documentation Interest Group (DIG) Monthly Meeting |
| 14:01 |
pinesol_green |
Meeting started Thu Mar 5 14:00:59 2015 US/Eastern. The chair is yboston. Information about MeetBot at http://wiki.debian.org/MeetBot. |
| 14:01 |
pinesol_green |
Useful Commands: #action #agreed #help #info #idea #link #topic. |
| 14:01 |
pinesol_green |
The meeting name has been set to '2015_03_05___dig_monthly_meeting_evergreen_documentation_interest_group__dig__monthly_meeting' |
| 14:01 |
yboston |
The agenda can be found here http://evergreen-ils.org/dokuwiki/doku.php?id=evergreen-docs:dig_meeting_20150305-agenda |
| 14:01 |
|
rsoulliere joined #evergreen |
| 14:01 |
yboston |
#topic Introductions |
| 14:01 |
yboston |
Please feel free to start introducing yourselves... |
| 14:02 |
remingtron |
#info remingtron is Remington Steed, Hekman Library (Calvin College) |
| 14:02 |
kbutler |
#info kbutler is Kate Butler, Rodgers Memorial Library |
| 14:02 |
yboston |
#info yboston is Yamil Suarez @ Berklee College of Music - DIG meeting facilitator |
| 14:02 |
sandbergja |
#info sandbergja is Jane Sandberg, Linn-Benton Community College |
| 14:03 |
rsoulliere |
#info rsoulliere is Robert Soulliere, Mohawk College |
| 14:03 |
yboston |
ok lets move on |
| 14:04 |
yboston |
any reports from coordinators (no woorries if there are none)? |
| 14:04 |
rsoulliere |
none from me. |
| 14:05 |
yboston |
OK |
| 14:05 |
yboston |
#topic old action items |
| 14:05 |
yboston |
#info 1) yboston will contact krvmga to check in on his 2.7 docs work |
| 14:05 |
yboston |
I spoke to krvmga, and apparently this has been taken care of |
| 14:06 |
yboston |
#info 2) kmlussier will complete the marc stream importer work and check on the status of the RDA docs |
| 14:06 |
remingtron |
yboston: question about #1 |
| 14:06 |
yboston |
go ahead |
| 14:06 |
remingtron |
does that mean it's been pushed to master? |
| 14:06 |
yboston |
that is an excelent question |
| 14:07 |
remingtron |
otherwise, we just need to know where his finished work is so we can review and push it in |
| 14:07 |
yboston |
the notes I had at the last time I contacted him did not specify |
| 14:07 |
yboston |
what was missing, only that I had to check in on him |
| 14:07 |
yboston |
I can ask him again to verify exactly what it was |
| 14:07 |
yboston |
kmlussier: do you remember what it was? |
| 14:08 |
yboston |
I thought it either had to do with added content or with the MARC stream importer? |
| 14:08 |
remingtron |
looks like added content |
| 14:09 |
remingtron |
#link http://evergreen-ils.org/dokuwiki/doku.php?id=evergreen-docs:2.7_needs |
| 14:09 |
remingtron |
(look for Jim Keenan, under OPAC section) |
| 14:09 |
yboston |
I see |
| 14:09 |
kmlussier |
Sorry, I stepped away. |
| 14:10 |
yboston |
I am grepping the docs to see what I find under "cafe" |
| 14:10 |
|
bmills joined #evergreen |
| 14:10 |
kmlussier |
I haven't seen Jim's documentation merged anywhere. However, there was some other work done on Added Content since then. I don't know if it overlaps or not. |
| 14:11 |
kmlussier |
I think Josh Stompro may have done that work? |
| 14:11 |
* kmlussier |
still needs to defer her action item. |
| 14:11 |
kmlussier |
Sorry sandbergja! I know I owe you some notes. |
| 14:12 |
sandbergja |
kmlussier: no worries! |
| 14:13 |
yboston |
It looks like there is nno recent updates to content related to "content cafe" in master right now |
| 14:13 |
yboston |
here is a link ot the history of the file that covers contnet cafe http://git.evergreen-ils.org/?p=Evergreen.git;a=history;f=docs/admin_initial_setup/designing_your_catalog.txt;h=4699ae2fe912acc71fbc7729bd8a8330d2c9b780;hb=master |
| 14:15 |
yboston |
I can follow up with krvmga to verify where the content is; or if he can provide with another copy of it |
| 14:15 |
remingtron |
I was expecting Jim to send us a link to his docs, then we would review and push it in |
| 14:15 |
remingtron |
so I figured since we didn't hear anything, we just needed to follow up with him |
| 14:16 |
yboston |
I will reach out to him again |
| 14:16 |
remingtron |
yboston++ |
| 14:16 |
yboston |
#action yboston will reach out to krvmga for another copy of his docs to push to master |
| 14:16 |
yboston |
#action kmlussier will complete the marc stream importer work and check on the status of the RDA docs |
| 14:16 |
yboston |
anything else before I move on to older action items? |
| 14:17 |
yboston |
will move on |
| 14:17 |
yboston |
#info 3) yboston will follow up with gmcharlt and Robert about provisioning a VM for DIG |
| 14:18 |
yboston |
I have spoken to gmcharlt , we hope to have a partially set up VM next week |
| 14:18 |
remingtron |
hurray! |
| 14:18 |
yboston |
though we would need to then start configuring it for docs processing |
| 14:19 |
yboston |
gmcharlt++ |
| 14:19 |
yboston |
csharp++ |
| 14:19 |
yboston |
any additional comments on this topic? |
| 14:19 |
yboston |
moving on |
| 14:19 |
yboston |
#info 4) remingtron will set up doodle to set up a meeting on IRC in February 2015 to discuss re-organizing the EG docs |
| 14:20 |
yboston |
done and done |
| 14:20 |
remingtron |
yup |
| 14:20 |
yboston |
remingtron++ |
| 14:20 |
yboston |
I thought it was great t have a whle meeting on one subject |
| 14:20 |
yboston |
though it might been a nightmare for others |
| 14:20 |
remingtron |
me too, very focused |
| 14:20 |
kbutler |
I thought it was great, even though I ended up missing most of it in person. |
| 14:21 |
yboston |
there were some next action from that meeting, none of which I currently recall |
| 14:21 |
yboston |
should I add them to this meeting too? |
| 14:21 |
remingtron |
yboston: we can discuss it under "New Business" |
| 14:21 |
remingtron |
I can remind you about some of them |
| 14:21 |
yboston |
OK, lets move on (thanks) |
| 14:21 |
yboston |
#info 5) all check in on state of latest web client docs from bshum or others |
| 14:22 |
remingtron |
we can get to that in a minute too, under Old Business, and I'll explain the state of things |
| 14:23 |
yboston |
I got a chance to install OSRF 2.4 and saw soem of the web clietn install docs, but I have not verified that the process worked |
| 14:23 |
yboston |
OK, moving on |
| 14:23 |
yboston |
#info 6) remingtron will set up a doodle poll and ask on the DIG list to set up a time for a DIG 2.8 hackfest |
| 14:23 |
|
bmills joined #evergreen |
| 14:23 |
yboston |
done & and to be done tomorrow |
| 14:23 |
yboston |
remingtron++ |
| 14:23 |
remingtron |
yes, excited for the hackfest |
| 14:24 |
yboston |
that is the last of the old business |
| 14:24 |
yboston |
sorry, the last of the previous meeting action items |
| 14:25 |
yboston |
any preference between tacklign old busniess versus new business? I rather tackle the hackfest first |
| 14:25 |
yboston |
(sorry for my bad spelling) |
| 14:25 |
remingtron |
sure, hackfest first |
| 14:26 |
yboston |
#topic DIG hackfest |
| 14:26 |
yboston |
thanks to kmlussier & remingtron for helping to move things forward |
| 14:26 |
yboston |
I can host the goolge hangout |
| 14:26 |
remingtron |
great! |
| 14:27 |
yboston |
any other questions or comments ont he hackfest? |
| 14:27 |
remingtron |
I can't join until about 9am EST |
| 14:28 |
yboston |
on a related note, I will be out for a meeting bewteen 9:50 and 11 AM EST |
| 14:28 |
remingtron |
can you leave the hangout open during that time? |
| 14:28 |
yboston |
yes |
| 14:28 |
yboston |
out of curiosity, can I get a show of hands of how can make it? |
| 14:29 |
yboston |
also, if possible I guess we can start markign what we want to work on |
| 14:29 |
yboston |
#link http://evergreen-ils.org/dokuwiki/doku.php?id=evergreen-docs:2.8_needs |
| 14:30 |
yboston |
now that I thik about it, since we are not meeting in groups in a physical location there might not be much to talk about right now |
| 14:30 |
yboston |
any final questosn or comments on the topic of the hackfest? |
| 14:30 |
remingtron |
just that all who are willing should sign up for something |
| 14:31 |
* kmlussier |
raises here hand |
| 14:31 |
remingtron |
the goal is to have all 2.8 features documented and pushed to master before 2.8 is released |
| 14:31 |
remingtron |
I think the target release date is still March 18 |
| 14:31 |
|
bmills joined #evergreen |
| 14:32 |
yboston |
kmlussier: did you have a question? |
| 14:32 |
yboston |
or just sowing support |
| 14:32 |
yboston |
*showing |
| 14:32 |
kmlussier |
yboston: You asked for a show of hands |
| 14:32 |
kmlussier |
So I showed my hand. :) |
| 14:32 |
yboston |
oops |
| 14:32 |
yboston |
thanks |
| 14:32 |
yboston |
kmlussier++\ |
| 14:32 |
yboston |
kmlussier++ |
| 14:33 |
yboston |
I can move on to other business now |
| 14:34 |
yboston |
let me stick to new busniess |
| 14:34 |
yboston |
#topic We should use .asciidoc file extension. |
| 14:34 |
remingtron |
this and the next topic were suggestions from.....someone I can't remember |
| 14:35 |
remingtron |
and I lost that IRC session. Josh someone? |
| 14:35 |
remingtron |
sorry whoever you are! |
| 14:35 |
yboston |
I can't remember either, but I really like having Github parse the docs automatically |
| 14:35 |
yboston |
Justin Hopkins? |
| 14:35 |
remingtron |
yes, I think that's it! |
| 14:36 |
kmlussier |
rsoulliere: Does that cause any problems with the doc build? |
| 14:36 |
remingtron |
hopkinsju++ |
| 14:37 |
rsoulliere |
kmlussier: shouldn't cause any problems. |
| 14:37 |
dbs |
kmlussier: all of the references to the doc source filenames would need to be adjusted, a one time pain |
| 14:37 |
|
bmills joined #evergreen |
| 14:37 |
remingtron |
dbs: right, we'd have to choose a time and make the switch |
| 14:38 |
remingtron |
so if everyone agrees, I can help coordinate the time, probably after 2.8 is released |
| 14:38 |
* kmlussier |
has no opinion on the matter |
| 14:38 |
kmlussier |
Just tell me what to do, and I'll do it. :) |
| 14:38 |
remingtron |
anyone have concerns? |
| 14:39 |
yboston |
I could argue that the use of the txt suffix looks friendlier to newcomers than what we are proposing |
| 14:39 |
yboston |
the files can open by default ona windows or Mac machine |
| 14:39 |
remingtron |
yboston: yes, but GitHub will be much friendlier by auto-generating the HTML to preview |
| 14:40 |
yboston |
remingtron: I agree, that part is very enticing |
| 14:40 |
remingtron |
true, we would lose the niceness for working on their local computer |
| 14:41 |
yboston |
we can create a guide to teach folks to use Notepad or Notepadd++ to open these files |
| 14:41 |
* dbs |
doesn't have a sense for how many people contribute to the docs via github |
| 14:41 |
yboston |
it could help promote the uses of gists even more |
| 14:42 |
remingtron |
I think we've had more github activity recently |
| 14:42 |
yboston |
should we have a vote on the mailing list? |
| 14:43 |
yboston |
(for the change) |
| 14:43 |
remingtron |
sure, I'll send a note to the list |
| 14:43 |
remingtron |
that will get more voices, hopefully |
| 14:43 |
yboston |
#action remingtron will send out emial to list about cahnging the file suffix in the docs repo to .asciidoc |
| 14:44 |
yboston |
#topic We should use the alternate style of headings |
| 14:44 |
yboston |
#link http://paste.evergreen-ils.org/36 |
| 14:44 |
* dbs |
thinks .adoc might work too, which has the advantage of being shorter |
| 14:45 |
* dbs |
confirms .adoc works : https://github.com/opendevise/asciidoc-samples |
| 14:45 |
kmlussier |
I've seen both styles of headings used in our docs. I'm not sure I see the benefit of choosing one over the other. Couldn't people just use whichever they prefer? |
| 14:45 |
remingtron |
dbs++ |
| 14:45 |
yboston |
dbs++ |
| 14:45 |
kbutler |
I think the alternate style is more readable unrendered. |
| 14:46 |
yboston |
kbutler: I agree |
| 14:46 |
remingtron |
kmlussier: the underline style is easy to get wrong, must be the exact right length |
| 14:46 |
remingtron |
kmlussier: and underlines are hard to remember which character to use for which level |
| 14:46 |
remingtron |
unintuitive |
| 14:46 |
kbutler |
remingtron: I agree on all points |
| 14:47 |
remingtron |
I would call them minor differences, but worth changing |
| 14:47 |
* dbs |
agrees that it comes down to preferences |
| 14:47 |
yboston |
kbutler: I misread, I disagree with you :( |
| 14:47 |
|
bmills joined #evergreen |
| 14:47 |
dbs |
and potentially more churn in two more years if a different group decides they prefer a different style |
| 14:47 |
remingtron |
dbs: hm, good point |
| 14:48 |
yboston |
I guess it would probably get confusing if we were using both styles all over the place |
| 14:48 |
kmlussier |
dbs: A different group? Where are we going? ;) |
| 14:48 |
|
mglass joined #evergreen |
| 14:49 |
remingtron |
well, it's certainly not an urgent issue, so let's move the discussion to the list |
| 14:49 |
yboston |
either way, we should turn to the list for more opinions / votes |
| 14:49 |
remingtron |
yboston: my thoughts exactly! |
| 14:49 |
yboston |
:) |
| 14:49 |
kmlussier |
My preference is to leave it so that people use whichever style they prefer. But, if others want to go with the alternate style, I'll learn the other way. |
| 14:49 |
yboston |
and we would need to pick a time to do the switch, if we want to |
| 14:50 |
yboston |
I am a bit concerend that it will ger really confusing if we mix two styles in the same file |
| 14:50 |
remingtron |
yboston: agreed, we should always use the same style within one file |
| 14:51 |
remingtron |
any edits or additions should use the existing style |
| 14:51 |
yboston |
does any one to volunteer to send an email to the list to talk abotu this further? |
| 14:51 |
remingtron |
I'll send that email to the list in a few weeks |
| 14:51 |
kmlussier |
That's fine with me. |
| 14:52 |
yboston |
#action remingtron will send an emial to the list to consider changing the AsciiDoc section header style |
| 14:52 |
yboston |
we have less than ten minutes |
| 14:52 |
|
bmills joined #evergreen |
| 14:52 |
yboston |
shoudl we cover web clients or the re-org? |
| 14:52 |
yboston |
I mean cover first |
| 14:52 |
remingtron |
one question about the re-org |
| 14:53 |
yboston |
go agead |
| 14:53 |
remingtron |
does everyone think we should have another special meeting about it? |
| 14:53 |
remingtron |
if so, I'll send another doodle to pick the time |
| 14:53 |
yboston |
I definetly want another meeting at some point |
| 14:54 |
remingtron |
otherwise, let's discuss it next meeting |
| 14:54 |
yboston |
though, I prefer that someone has summarized the notes from the first meeting first |
| 14:54 |
remingtron |
there's a few links here: |
| 14:54 |
remingtron |
http://evergreen-ils.org/dokuwiki/doku.php?id=evergreen-docs:reorg_2014 |
| 14:54 |
remingtron |
I guess a few of the links are empty pages |
| 14:55 |
remingtron |
but I posted my ideas here: |
| 14:55 |
remingtron |
http://evergreen-ils.org/dokuwiki/doku.php?id=evergreen-docs:reorg_2014:audience_focused_layout |
| 14:55 |
remingtron |
about presenting the docs as separate books related to user type |
| 14:55 |
yboston |
remingtron++ |
| 14:56 |
kbutler |
remingtron++ |
| 14:57 |
yboston |
we can wait untilt he next meeting to talk abotu this again and decide whn to have another meeitng |
| 14:57 |
yboston |
or I can create a doodle poll for another meeting |
| 14:57 |
remingtron |
sure, let's wait until next meeting |
| 14:57 |
yboston |
ok |
| 14:58 |
remingtron |
and all ideas are welcome! you can edit those wiki pages! |
| 14:58 |
remingtron |
(you = anyone) |
| 14:58 |
yboston |
#action all decide how to proceed with docs reorganization next meeting |
| 14:58 |
yboston |
we have two mintues left |
| 14:58 |
yboston |
any questions or comemnts about web cleitn docs? |
| 14:59 |
yboston |
*client |
| 14:59 |
remingtron |
just web client |
| 14:59 |
remingtron |
oh, right |
| 14:59 |
remingtron |
one comment |
| 14:59 |
remingtron |
there are 3 docs ready to review on this page: |
| 14:59 |
remingtron |
http://evergreen-ils.org/dokuwiki/doku.php?id=evergreen-docs:webclient |
| 15:00 |
remingtron |
we should try to review and push them to master. no rush, just didn't want to forget them |
| 15:00 |
yboston |
remingtron++ |
| 15:00 |
remingtron |
that's all |
| 15:00 |
yboston |
anything else? |
| 15:00 |
yboston |
from enyone else? |
| 15:01 |
yboston |
OK then, thanks everyone and hope to see soem of you tomorrow |
| 15:01 |
yboston |
#endmeeting |
| 15:01 |
pinesol_green |
Meeting ended Thu Mar 5 15:01:22 2015 US/Eastern. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) |
| 15:01 |
pinesol_green |
Minutes: http://evergreen-ils.org/meetings/evergreen/2015/evergreen.2015-03-05-14.00.html |
| 15:01 |
pinesol_green |
Minutes (text): http://evergreen-ils.org/meetings/evergreen/2015/evergreen.2015-03-05-14.00.txt |
| 15:01 |
pinesol_green |
Log: http://evergreen-ils.org/meetings/evergreen/2015/evergreen.2015-03-05-14.00.log.html |
| 15:01 |
remingtron |
yboston++ |
| 15:01 |
kmlussier |
yboston++ |
| 15:01 |
kbutler |
yboston++ |
| 15:06 |
remingtron |
yboston: did gmcharlt mention to you an epub error? |
| 15:06 |
yboston |
remingtron: yes |
| 15:06 |
remingtron |
did you fix it? works fine for me |
| 15:07 |
gmcharlt |
I subsequent spoke with rsoulliere |
| 15:07 |
gmcharlt |
he fixed it |
| 15:07 |
gmcharlt |
rsoulliere++ |
| 15:07 |
remingtron |
rsoulliere++ |
| 15:07 |
remingtron |
then we can mark bug 1428780 as fixed! |
| 15:07 |
pinesol_green |
Launchpad bug 1428780 in Evergreen "Documentation 404" (affected: 1, heat: 6) [Undecided,New] https://launchpad.net/bugs/1428780 |
| 15:07 |
yboston |
rsoulliere++ |
| 15:07 |
rsoulliere |
I fixed locally, not in repo yet. |
| 15:08 |
remingtron |
ah, what was the cause? |
| 15:08 |
rsoulliere |
remingtron: a2x epub processing was breaking on an empty section (no paragraph under section heading) in the docs. |
| 15:09 |
remingtron |
rsoulliere: do you have time to fix the asciidoc? or want someone else to look at it? |
| 15:10 |
|
bmills joined #evergreen |
| 15:16 |
rsoulliere |
I asked yboston and kmlussier to fix in the repo. I will not have time today. |
| 15:19 |
|
RoganH joined #evergreen |
| 15:20 |
Dyrcona |
What's the recommended way to set collected and reacting events back to pending so they will run again? |
| 15:20 |
Dyrcona |
I've got about 100 from yesterday afternoon that I want to reset. |
| 15:21 |
csharp |
Dyrcona: this is what I did when we saw notices failing when we activated longoverdue: http://git.evergreen-ils.org/?p=contrib/pines.git;a=blob;f=longoverdue/batch_process_longoverdue_notices.sh;h=77dfe492ad05264d3d5d360332596168e8802861;hb=HEAD |
| 15:23 |
Dyrcona |
Thanks, csharp. I was not sure what dates and other fields to mess with. |
| 15:23 |
csharp |
it took a while for me to figure those out too |
| 15:24 |
* csharp |
has to rely on this script because he can't remember :-) |
| 15:25 |
Dyrcona |
I know what you mean about relying on scripts. |
| 15:27 |
Dyrcona |
Heh. Looks like I only have 50 from yesterday. |
| 15:27 |
Dyrcona |
The other 50 are from Boxing Day. |
| 15:29 |
Dyrcona |
If I set them back to pending the normal --run-pending should pick them up, no? |
| 15:29 |
Dyrcona |
guess I can wait a few minutes and look. |
| 15:29 |
csharp |
as long as the delay/max_delay allow it to go back that far |
| 15:31 |
Dyrcona |
csharp++ |
| 15:32 |
Dyrcona |
And, where is that set? tsbere usually messes with out A/T stuff, though I've mangled a few of the templates. |
| 15:32 |
csharp |
in the DB, it's in action_trigger.event_definition |
| 15:33 |
csharp |
In the SC Admin -> Local Admin -> Notifications/Action Triggers & double-click the right row |
| 15:34 |
csharp |
the resulting SQL will be "BETWEEN delay AND max_delay" basically |
| 15:34 |
Dyrcona |
csharp: Thanks, I was suspecting it would be there. |
| 15:35 |
csharp |
hth |
| 15:35 |
Dyrcona |
These are all event defs 5 and 103, email and sms hold notices. |
| 15:35 |
Dyrcona |
I believe it does. Thanks muchly! |
| 15:35 |
Dyrcona |
And I should turn sql history back on in PGAdmin, on a slightly related note. |
| 15:35 |
csharp |
having spent all day myself in those event defs, it's all pretty fresh in my mind ;-) |
| 15:36 |
Dyrcona |
I turned it off for various reasons. |
| 15:36 |
Dyrcona |
Yeah. |
| 15:39 |
Dyrcona |
Delay interval looks like 30 minutes and max delay interval is null, so guess that means forever? |
| 15:39 |
csharp |
yes, iirc |
| 15:40 |
Dyrcona |
Guess, I'll have to give it another ten minutes or so. Our next run pending will happen at a quarter to four. |
| 15:56 |
Dyrcona |
And it worked. |
| 15:59 |
Dyrcona |
csharp: You might want to null run_time, too. I notice it doesn't change if already set. |
| 16:06 |
Dyrcona |
What a day! |
| 16:06 |
Dyrcona |
In private conversation, it sounds like Bibliomation may have bumped into what we're encountering, but they never looked into as deeply. |
| 16:07 |
Dyrcona |
They made a work around by splitting out even more granularities for overdue and pre-overdue notices. |
| 16:07 |
Dyrcona |
I want to see if I can fix the cause. |
| 16:08 |
Dyrcona |
heh. |
| 16:09 |
* Dyrcona |
finds it amusing when I have 55 line queries where 50 of the lines are copy/pasted ids or barcodes. |
| 16:09 |
* Dyrcona |
has one from earlier this week that about 8,800 lines long, most of it barcodes. |
| 16:09 |
Dyrcona |
my fingers have gotten in my way all day, too. |
| 16:12 |
Dyrcona |
@monologue |
| 16:12 |
pinesol_green |
Dyrcona: Your current monologue is at least 12 lines long. |
| 16:13 |
Dyrcona |
Anyone ever delete old/obsolete events from action_trigger.event? |
| 16:13 |
Dyrcona |
We have like 20 million of them. |
| 16:20 |
|
gdunbar joined #evergreen |
| 16:21 |
|
kmlussier_at_the joined #evergreen |
| 16:21 |
eeevil |
Dyrcona: you can trash old rows if you don't need them anymore. I don't recall if the delete will cascade to the output table, though |
| 16:22 |
jboyer-isl |
Dyrcona: We've thought about it, but didn't want to try to determine which events would be auto-generated later. I may be mistaken in that this is even possible, though. |
| 16:22 |
* jboyer-isl |
needs to look into more max delay age stuff before doing that. |
| 16:22 |
berick |
jboyer-isl: it's possible in some cases |
| 16:23 |
Dyrcona |
Thanks, guys. I'll look into some tomorrow. If there's a cascade, it should be pretty obvious. |
| 16:23 |
* berick |
says mysteriously |
| 16:23 |
eeevil |
Dyrcona: oh, yeah, beware what jboyer-isl was referring to for passive events |
| 16:23 |
jboyer-isl |
berick: Ah, and my fears are thus founded. ;) |
| 16:24 |
Dyrcona |
Ok. Can't delete passive events. |
| 16:24 |
Dyrcona |
I assume that's like the overdue to lost type of things? |
| 16:24 |
eeevil |
well, you might be able to, if they're past max_delay (assuming it's set) |
| 16:24 |
eeevil |
right |
| 16:24 |
jboyer-isl |
Is that the difference between active and passive, active events are created ahead of time and run when their run_time hits, and passive are generated externally? |
| 16:24 |
eeevil |
or even just overdue notices |
| 16:24 |
|
pmurray joined #evergreen |
| 16:24 |
Dyrcona |
Ok. Good to know. |
| 16:25 |
eeevil |
jboyer-isl: sorta ... active event rows are inserted when the triggering even occurs. passive event rows are inserted when some timestamp is a certain age old (where "old" can be negative; see: pre-due notices) |
| 16:26 |
eeevil |
IOW, active events do not depend on a context time -- they just happen |
| 16:26 |
jboyer-isl |
eeevil: That makes sense. Enough to think about looking into things at some future date anyway. |
| 16:26 |
eeevil |
passive events are always in relation to a timestamp column on the target object |
| 16:27 |
jboyer-isl |
So the only way to prevent passive event regeneration is to set max_delay_age or remove it's head? |
| 16:27 |
|
julialima_ left #evergreen |
| 16:27 |
eeevil |
yes |
| 16:28 |
jboyer-isl |
Sounds like an A/T audit is in my future. One of the many ideas I have that I lack the time to act on. :/ |
| 16:28 |
jboyer-isl |
thanks |
| 16:28 |
jboyer-isl |
eeevil++ berick++ |
| 16:28 |
dbwells |
In many cases the regenerated passive event will no longer pass validation anyway, so there is at least some level of protection there. |
| 16:29 |
Dyrcona |
Yeah, I noticed that a number of the hold notifications failed validation when rerun today, but there was no error output to look into it. |
| 16:29 |
jboyer-isl |
dbwells: That's true, especially for any holds notification type events. |
| 16:30 |
Dyrcona |
I assume most of these were picked up/no longer on hold for the patron. |
| 16:31 |
|
Newziky left #evergreen |
| 16:49 |
|
bmills joined #evergreen |
| 17:02 |
|
mrpeters left #evergreen |
| 17:19 |
|
mmorgan left #evergreen |
| 19:39 |
|
bmills1 joined #evergreen |
