| Time |
Nick |
Message |
| 01:19 |
|
bwicksall joined #evergreen |
| 06:57 |
|
agoben joined #evergreen |
| 07:06 |
|
rjackson_isl joined #evergreen |
| 07:24 |
|
collum joined #evergreen |
| 08:00 |
|
tlittle joined #evergreen |
| 08:20 |
|
Dyrcona joined #evergreen |
| 08:21 |
remingtron |
Bmagic++ |
| 08:21 |
remingtron |
devted++ |
| 08:24 |
csharp |
all_yall++ |
| 08:25 |
agoben |
Heh, I'm with csharp on that. Very awesome to see all the work that was accomplished this week! |
| 08:26 |
csharp |
agoben++ # especially |
| 08:26 |
JBoyer |
agoben++ |
| 08:26 |
tlittle |
Yes, definitely agoben++ |
| 08:27 |
remingtron |
agoben++ #great hostess |
| 08:28 |
agoben |
Aww, thanks! Always a pleasure to have everyone here! |
| 08:29 |
csharp |
also... |
| 08:29 |
csharp |
agobens_mom++ |
| 08:30 |
csharp |
pretty sure sugarcream pie is going to be on our thanksgiving table :-) |
| 08:31 |
|
bos20k joined #evergreen |
| 08:35 |
csharp |
oh - nice - FF 70 with Enhanced Tracking Protection - I got a notification that it blocked Facebook trackers on a site I was just on |
| 08:35 |
csharp |
firefox++ |
| 08:35 |
csharp |
mozilla++ |
| 08:37 |
Dyrcona |
agoben++ |
| 08:40 |
collum |
Ditto - agoben++ |
| 08:41 |
Dyrcona |
GB2312_data_that_is_really_KOI8-- |
| 08:41 |
Dyrcona |
try/except++ |
| 08:42 |
Dyrcona |
...since we're doling out the karma. :) |
| 08:42 |
Dyrcona |
chardet++ |
| 08:42 |
|
mmorgan joined #evergreen |
| 08:45 |
Dyrcona |
Email can be worse than MARC when it comes to character set issues. |
| 08:48 |
dbwells |
agoben++ |
| 08:48 |
* Dyrcona |
goes back to writing an email to summarize hack-away activity for CW MARS staff. |
| 08:53 |
* mmorgan |
would love to see a message summarizing the hackaway go out to the general list. |
| 08:57 |
dbwells |
"...It was good, but not as good as when Dyrcona is there..." |
| 08:57 |
Dyrcona |
Ha! |
| 09:02 |
Dyrcona |
I'm sorry I missed the sugar cream pie. |
| 09:02 |
Dyrcona |
:) |
| 09:03 |
* Dyrcona |
has had a lot going on lately and didn't really have the time to go this year. |
| 09:04 |
|
mdriscoll joined #evergreen |
| 09:04 |
Dyrcona |
mmorgan: https://wiki.evergreen-ils.org/doku.php?id=hack-a-way:hack-a-way-2019-agenda |
| 09:15 |
mmorgan |
Dyrcona: Thanks, I was watching that. Was just thinking that something put out to the list for general consumption might bring to the attention of the general community how much gets accomplished. |
| 09:18 |
Dyrcona |
Well, someone who was there should probably do that one. :) |
| 09:19 |
mmorgan |
Right! |
| 09:27 |
|
yboston joined #evergreen |
| 09:55 |
|
mmorgan left #evergreen |
| 09:57 |
|
sandbergja joined #evergreen |
| 10:02 |
|
mmorgan joined #evergreen |
| 10:10 |
|
nfBurton joined #evergreen |
| 10:27 |
|
collum_ joined #evergreen |
| 10:28 |
|
collum_ joined #evergreen |
| 10:29 |
|
collum_ joined #evergreen |
| 10:30 |
|
collum joined #evergreen |
| 10:31 |
|
collum joined #evergreen |
| 10:32 |
|
collum__ joined #evergreen |
| 10:35 |
|
collum_ joined #evergreen |
| 10:44 |
dbs |
huh, the links for the qatests failures return a 404, is that a known issue? |
| 10:48 |
|
collum joined #evergreen |
| 10:48 |
|
sandbergja joined #evergreen |
| 10:50 |
berick |
dbs: i think that happens if a test is in progress |
| 10:51 |
|
collum joined #evergreen |
| 10:56 |
dbs |
berick: ahh |
| 11:00 |
|
jvwoolf joined #evergreen |
| 11:01 |
pinesol |
News from qatests: Failed Installing AngularJS web client <http://testing.evergreen-ils.org/~live/test.28.html#2019-10-24T11:00:47,985113758-0400 -0> |
| 11:01 |
pinesol |
News from qatests: Failed Installing Angular web client <http://testing.evergreen-ils.org/~live/test.29.html#2019-10-24T11:00:48,026755118-0400 -2> |
| 11:01 |
pinesol |
News from qatests: Failed Log Output: osrfsys.log <http://testing.evergreen-ils.org/~live/test.76.html#2019-10-24T11:00:48,068169737-0400 -4> |
| 11:01 |
|
collum_ joined #evergreen |
| 11:02 |
|
collum__ joined #evergreen |
| 11:02 |
|
Christineb joined #evergreen |
| 11:02 |
berick |
I have a fix for the Angular test failures if someone could S/O and merge... |
| 11:03 |
berick |
working/user/berick/lp1830973-ang8-test-spec-repair |
| 11:03 |
berick |
note the LP number references the (now-closed) angular 8 bug |
| 11:19 |
Bmagic |
agoben++ |
| 11:26 |
stephengwills |
I’m converting MARC21 to MARCXML w/ the python script and hitting “UnicodeDecodeError: 'marc8_to_unicode' codec can't decode bytes in position 0-156: invalid multibyte character encoding”. is there a way to tell that script to skip the record it’s barfing on and keep processing the file, or is that a bad idea? |
| 11:26 |
jeff |
"the python script"? |
| 11:26 |
stephengwills |
http://docs.evergreen-ils.org/3.2/_migrating_your_bibliographic_records.html |
| 11:26 |
jeff |
pymarc had some annoying charset issues last time I tried to make use of it a few years ago. |
| 11:27 |
jeff |
I'm not sure the origins of that script. |
| 11:28 |
stephengwills |
I was hoping not to have to hack pymarc :) |
| 11:28 |
jeff |
Those docs might need some attention. I'm not sure how close to reality they are. |
| 11:29 |
Dyrcona |
abbreviations_in_configuration-- # I'm looking at you, haproxy's option httpchk. |
| 11:32 |
berick |
Status 220 HTTP Chicken |
| 11:35 |
JBoyer |
berick++ |
| 11:35 |
berick |
i'm suddenly craving Beef Manhattan |
| 11:36 |
Dyrcona |
:) |
| 11:37 |
Bmagic |
I am finding that some days SIP is not running. The last log entry is: raw_transport: shutting down: EOF during login |
| 11:37 |
Dyrcona |
Well, the httpchk option isn't doing what I want. |
| 11:39 |
Dyrcona |
Bmagic: I've not seen that. |
| 11:39 |
Bmagic |
strange |
| 11:40 |
Dyrcona |
I want haproxy to mark a brick down if OpenSRF services go down and apache starts returning 500 errors. I know there's a way to do this, and I'll figure it out, eventually. |
| 11:41 |
Dyrcona |
Bmagic: SIPServer should fork. It might shut the main process down if max_servers is not set or equals 1. |
| 11:42 |
* dbs |
was the origin of that script, jeff / stephengwills |
| 11:42 |
Dyrcona |
But I can't think of anything else that would do that. |
| 11:43 |
Bmagic |
this machine is setup PreFork and max_servers='200' |
| 11:44 |
dbs |
stephengwills: pymarc does not fail gracefully when it encounters encoding issues - often either because LDR09 is lying, or because non-UTF8 / non-MARC8 chars (like iso-8859-1) have been added by other less strict tools |
| 11:44 |
berick |
Bmagic: could be a client connecting to the socket, but never sending the login message (within the configured timeout). |
| 11:45 |
berick |
hm, though, I would also expect to see a timeout log message in that case |
| 11:45 |
berick |
or the client opens the socket, then disconnects without ever doing anything. |
| 11:47 |
Dyrcona |
Sure, but that shouldn't kill all of the SIPServer processes. |
| 11:48 |
berick |
oh, yeah, it certainly should not. I missed that bit. |
| 11:48 |
Bmagic |
maybe they die off one at a time throughout the day? |
| 11:48 |
Bmagic |
(and never come back....) |
| 11:49 |
dbs |
stephengwills: so as long as your MARC records are perfectly encoded and structured, pymarc should be happy - heh |
| 11:49 |
stephengwills |
dbs record.out should hold, up to the record containing the offending char and allow me to correct the offending bib, I assume, and then I rinse and repeat? |
| 11:50 |
Dyrcona |
I had SIPServer die mysteriously on one of our hosts recently, but never figured out why. |
| 11:50 |
stephengwills |
sorry…up to the record BEFORE the offending record |
| 11:50 |
* Dyrcona |
runs out to get some lunch. Back in a few. |
| 11:52 |
stephengwills |
seems like the next step is make this script try and catch or something. I’ll play with it. |
| 11:53 |
dbs |
stephengwills: a more forgving MARC21 to MARCXML tool is yaz-marcdump; something like "yaz-marcdump -i marc -o marcxml -f utf8 -t utf8 -l 9=97 inputfile.mrc > outputfile.xml" |
| 11:53 |
stephengwills |
ok… will try that first :). thanks |
| 11:54 |
dbs |
there was a recent discussion about "Encoding issues of Doom" on the pymarc mailing list, possibly a similar sitation to what you're facing: https://groups.google.com/d/msg/pymarc/grZpM5Me34s/BTQX6bnFBwAJ |
| 11:54 |
stephengwills |
ls |
| 11:55 |
stephengwills |
sorry |
| 11:57 |
stephengwills |
yaz-marcdump cleared it. Thanks Dan |
| 11:57 |
stephengwills |
I’ll add that to the docs. |
| 11:59 |
stephengwills |
now I wonder what Evergreen is going to do with those records when I try and ingest them :) |
| 12:05 |
dbs |
stephengwills: yeah, yaz-marcdump might have created some unusual MARC :) |
| 12:07 |
stephengwills |
as well as a lot of extra \n I’m needing to remove to get one record per line. but that just makes it all more fun. |
| 12:09 |
|
jihpringle joined #evergreen |
| 12:21 |
|
yboston joined #evergreen |
| 12:35 |
|
collum joined #evergreen |
| 12:36 |
* Dyrcona |
isn't he only one having encoding issues today, I see. |
| 12:38 |
|
alynn26 joined #evergreen |
| 12:46 |
Dyrcona |
Heh. "Fixed" my httpchk issue, but now they're all down because I forgot that / does a redirect. :) |
| 12:51 |
|
khuckins joined #evergreen |
| 12:52 |
Dyrcona |
Do websockets return http status codes? (maybe I should look that up.) |
| 12:57 |
JBoyer |
Dyrcona, maybe the upgrade response? Not sure if that's the initial response or only in response to certain types of request. |
| 12:58 |
Dyrcona |
JBoyer: Thanks, but I found a simpler solution. I can tell haproxy to check a different port or even a different server, so I think if I tell it to check port 80, then websockets is down when http is down, which is good enough. |
| 12:58 |
* Dyrcona |
is about to test that. |
| 12:59 |
JBoyer |
That's how I've always done it in ldirector, yeah. :) |
| 13:03 |
Dyrcona |
Yeah, it's working. |
| 13:03 |
Dyrcona |
:) |
| 13:04 |
jeff |
Assuming I followed that correctly, do you now run the danger of sending websockets requests to a backend where websockets are dead but HTTP on port 80 is fine? |
| 13:05 |
Dyrcona |
jeff: I will check that after I see if things come back automatically, but that said, I've never had websocketd die. |
| 13:07 |
Dyrcona |
Restart services and apache2 and the brick is working again. |
| 13:08 |
Dyrcona |
stop websocketd and it still shows as up. I really want to check both conditions.... |
| 13:13 |
Dyrcona |
Hm... Looks like I need to write an external command to do this. |
| 13:14 |
Dyrcona |
That's beyond the scope of the remainder of today and tomorrow. |
| 13:16 |
JBoyer |
Looks like page 4 of the RFC should help. You send a request with an upgrade header, you should get back a response with a 101 status code. |
| 13:16 |
JBoyer |
https://tools.ietf.org/html/rfc6455#section-1.2 |
| 13:17 |
JBoyer |
For whenever. :) |
| 13:21 |
Dyrcona |
JBoyer: Yes. I see that, and I've found some examples. A straightforward "check" works if websocketd is dead. I basically want to check that, plus if services have died, and I think I get the 101 even if services are dead, but I'll write a little script to check that. |
| 13:21 |
Dyrcona |
JBoyer++ |
| 13:21 |
Dyrcona |
jeff++ |
| 13:22 |
Dyrcona |
complication-- |
| 13:22 |
Dyrcona |
webosckets-- |
| 13:22 |
jeff |
habanero_hellfire_chili++ |
| 13:22 |
JBoyer |
Oh, yeah, websockets will outlive a service problem since it doesn't know what's going on, just where to throw the bytes. |
| 13:23 |
* JBoyer |
is almost, but not quite, jealous of jeff's chili. |
| 13:24 |
* Dyrcona |
had supermarket pepperoni pizza....is also jealous of jeff's chili. |
| 13:25 |
Dyrcona |
JBoyer: Yeah, but I just realized that writing this check script will be easier than I thought, so I might as well do it. |
| 13:25 |
JBoyer |
Nice. |
| 13:26 |
Dyrcona |
The question is: curl or LWP? |
| 13:26 |
Dyrcona |
@decide curl or LWP |
| 13:26 |
pinesol |
Dyrcona: go with curl |
| 13:50 |
remingtron |
Bmagic: your Antora README says "git clone git://git.evergreen-ils.org/Evergreen.git". Shouldn't that be "/working/Evergreen.git" ? |
| 13:50 |
remingtron |
meaning, I think we need "/working" added before "/Evergreen.git" |
| 14:01 |
Bmagic |
sure! |
| 14:01 |
Bmagic |
Pushing the change now |
| 14:01 |
Bmagic |
remingtron ^^ |
| 14:04 |
remingtron |
Bmagic: cool! |
| 14:05 |
remingtron |
I'm about to email the DIG list about our progress, and to invite helpers |
| 14:05 |
Bmagic |
remingtron: BTW - I've been pushing lots of changes this morning and last night as late as 1am, lol |
| 14:05 |
remingtron |
ha! nice work man |
| 14:06 |
Bmagic |
you GOTTA see the UI now |
| 14:07 |
Bmagic |
also - I saw you posted on LP about the next/last links. Those are easy to add with the handlebars lingo: https://docs.antora.org/antora-ui-default/templates/ |
| 14:08 |
|
sandbergja joined #evergreen |
| 14:08 |
Bmagic |
if we want that - I believe I can put it in there. {{page.next}} {{page.previous}} |
| 14:14 |
nfBurton |
How much of the OPAC UI is getting changed with this Angular update and how much is staying tt2? |
| 14:17 |
JBoyer |
nfBurton, I think the only Angular OPAC work is staff-only, so nothing changes for the public interface for now. |
| 14:18 |
nfBurton |
Oh okay |
| 14:18 |
nfBurton |
I was woried my work would be rendered pointless lol |
| 14:19 |
nfBurton |
I'm real excited for those speedy searches though! |
| 14:19 |
remingtron |
Bmagic++ #finding solution to prev/next links |
| 14:20 |
Bmagic |
remingtron: lol, just about done implementing it |
| 14:21 |
remingtron |
awesome |
| 14:22 |
nfBurton |
The working repo seems to be missing some npm modules? I haven't been able to build for the staff client. Is there a step |
| 14:22 |
nfBurton |
I am missing? |
| 14:23 |
jeff |
We've been using the Windows build of Hatch 0.3.2 from Comment 1 in bug 1830391. That installs without error on a 32 bit install of Windows, but then you have a Native Messaging Host that doesn't actually work (and "hatch.bat test" fails). |
| 14:23 |
pinesol |
Launchpad bug 1830391 in Evergreen "Hatch omnibus circa 3.3 (Java updates and more)" [Undecided,New] https://launchpad.net/bugs/1830391 |
| 14:23 |
jeff |
I suspect that I can drop a 32 bit build of OpenJDK in place and it'll then work, but I haven't tested that theory. |
| 14:23 |
Dyrcona |
Ok. So, I can connect to the websocket port, but it just hangs. If I do a HEAD request, I get a 405, which may be good enough. |
| 14:38 |
berick |
jeff: let me know how it goes... |
| 14:38 |
berick |
perhaps we'll need 2 windows installers |
| 14:44 |
jeff |
...and perhaps some installer logic to error out when trying to install 64 bit on 32 bit OS install. |
| 14:45 |
jeff |
one installer that's approximately twice the size probably wouldn't be ideal. |
| 14:45 |
jeff |
especially since the number of 32 bit installs would be (hopefully) low. |
| 14:47 |
remingtron |
nfBurton: what's going wrong? are you seeing errors? |
| 14:51 |
JBoyer |
jeff, I think 2 separate installers would be good, and erroring out on the wrong bit-ness is much easier than all of the back and forth we're currently doing. |
| 14:53 |
nfBurton |
yeah. most of the modules in .staging just aren't there |
| 14:54 |
nfBurton |
Open-ILS/src/eg2/node_modules/.staging |
| 14:54 |
nfBurton |
It expects them though :S |
| 14:56 |
|
yboston joined #evergreen |
| 14:56 |
JBoyer |
Oh, that. nfBurton, you might have to sudo rm -rf /usr/local/lib/node_modules and start again. May also have to clear out /usr/local/bin/ node, npm, ng, and npx |
| 14:57 |
JBoyer |
I was hitting that a lot earlier this week and that's what it took to get past it. :/ |
| 14:57 |
nfBurton |
oh! okay! |
| 14:57 |
nfBurton |
JBoyer++ |
| 14:58 |
JBoyer |
Also, bear in mind that if you run Makefile.install <osname>-developer you'll be getting a lot more packages. |
| 14:59 |
JBoyer |
(depending on the branch you |
| 14:59 |
JBoyer |
re using) |
| 15:01 |
Dyrcona |
Y'know, I usually like chroot, but not so much today. |
| 15:01 |
|
mmorgan1 joined #evergreen |
| 15:01 |
nfBurton |
yeah im using working and gathering all the dependencies |
| 15:02 |
Dyrcona |
osname-developer doesn't install that much. -packager or -translator install a pile of extra stuff. |
| 15:02 |
nfBurton |
Also, for anyone using the working branch, does the main My Account page work for you? I seem to get a 500 no matter what, even using the default template. |
| 15:04 |
nfBurton |
For my sanity |
| 15:04 |
jeff |
which working branch? |
| 15:05 |
nfBurton |
the main working branch |
| 15:06 |
nfBurton |
It seems the page with payments/charges wont show. |
| 15:08 |
jeff |
do you mean master? i'm not understanding "the main working branch", as the way we use working branches there are many... but perhaps confusingly the working repo also has a copy of master... |
| 15:08 |
jeff |
nfBurton: what is the commit hash of the tip of the branch you're asking about? |
| 15:08 |
Dyrcona |
I was about to ask the very same question. |
| 15:08 |
nfBurton |
I just cloned git://git.evergreen-ils.org/working/Evergreen.git |
| 15:08 |
|
jihpringle joined #evergreen |
| 15:08 |
jeff |
there's a main repo and a working repo, and many "working" branches in the working repo which mostly start with either user/username/ or collab/username |
| 15:09 |
nfBurton |
I havent actually pushed anything yet |
| 15:09 |
nfBurton |
Kind of a GIT noob. |
| 15:10 |
jeff |
got it, then if you just cloned from that URL your git client likely checked out a copy of the master branch from the working/Evergreen.git repo, which is normally automatically sync'd with the master branch of the git://git.evergreen-ils.org/Evergreen.git repo. |
| 15:11 |
nfBurton |
oh yes the master branch sorry |
| 15:11 |
jeff |
sorry, trying to be precise without sounding picky. |
| 15:12 |
Dyrcona |
Details matter. Programmers have to be picky. |
| 15:12 |
nfBurton |
No, it's fine. I'm still learning my way around this. And I'm like a day or 2 from pushing LP1778972. I just want to get it right |
| 15:12 |
jeff |
but back to my other question... what is the commit hash of the tip of the branch you're asking about? it would be the first line starting with "commit" in the output of "git show" or "git log" or... |
| 15:13 |
nfBurton |
Dyrcona++ |
| 15:13 |
nfBurton |
for sure |
| 15:13 |
jeff |
pinesol: bug 1778972 |
| 15:13 |
pinesol |
jeff: git diff origin/hamster Fleshing children complete |
| 15:13 |
pinesol |
Launchpad bug 1778972 in Evergreen "Wishlist - OPAC Redesign" [Wishlist,In progress] https://launchpad.net/bugs/1778972 - Assigned to Christopher Burton (cburton) |
| 15:14 |
nfBurton |
11868e97844a47d57489062197ac6809d9473366 |
| 15:14 |
pinesol |
nfBurton: [evergreen|Jason Boyer] Stamp upgrade script for LP 1849152 - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=11868e9> |
| 15:20 |
jeff |
nfBurton: okay, that matches current master. anything in apache logs when you get the 500 error? |
| 15:23 |
JBoyer |
Dyrcona, I'm just noticing your "-developer doesn't install that much..." I added chromium and firefox to that recently, though now that we're discussing it, maybe there should be a -tester target for those... (they're for replacing PhantomJS) |
| 15:24 |
Dyrcona |
Yes, there should probably be a -tester or -testing target. |
| 15:24 |
Dyrcona |
Or, they could go under -packager, since that's about the only time tests are run other than on the test server. |
| 15:26 |
JBoyer |
I suppose we could also expand that target out to include things like pgtap and other things... |
| 15:26 |
dbs |
Aren't commit reviewers / signers-off supposed to run tests? |
| 15:27 |
dbs |
I guess if we had a travis-ci environment or the like that tested every branch every time they were pushed, that would help alleviate the burden of running tests |
| 15:27 |
jeff |
the machine i do dev on isn't always the machine i run tests on. |
| 15:27 |
jeff |
but yes, better/easier CI is also a nice goal. |
| 15:28 |
Dyrcona |
dbs: I really doubt they do every time, and most new bugs are not caught by the tests. We've also been lax in adding tests. |
| 15:29 |
Dyrcona |
gitlab has CI. github has CI. |
| 15:30 |
Dyrcona |
I imagine CI for us would be not so much fun to set up, given how eclectic our codebase is. |
| 15:31 |
nfBurton |
@jeff egweb: Context Loader error: Can't locate object method "content" via package "OpenSRF::DomainObject::oilsMethodException" |
| 15:31 |
pinesol |
nfBurton: No, you're a puzzleheaded kraken! |
| 15:32 |
Dyrcona |
pinesol: Pipe down. It's hard enough to think with this headache. |
| 15:32 |
pinesol |
Dyrcona: Have you run autogen.sh? |
| 15:33 |
Dyrcona |
nfBurton: Have you run autogen.sh? |
| 15:33 |
nfBurton |
lol |
| 15:33 |
Dyrcona |
Sometimes the bot has good ideas. |
| 15:33 |
nfBurton |
I'm currently rebuilding but will |
| 15:34 |
Dyrcona |
I need to sort out what I'm going to do tomorrow since I spent today doing something unplanned and only partly successful. |
| 15:36 |
|
sandbergja joined #evergreen |
| 15:39 |
nfBurton |
still nothing. It seems to be looking for opac/api/mylist/retrieve.tt2 which doesn't exist |
| 15:42 |
nfBurton |
creating the file doesn't help though |
| 15:42 |
nfBurton |
Context Loader error: Can't locate object method "content" via package "OpenSRF::DomainObject::oilsMethodException" at /usr/local/share/perl/5.22.1/OpenILS/WWW/EGCatLoader/Account.pm line 2081. |
| 15:49 |
Dyrcona |
nfBurton: I haven't tried installing master in a few days. I was going to but got sidetracked. |
| 15:58 |
Bmagic |
remingtron: We have Previous/Next pushed to the branch |
| 15:59 |
remingtron |
Bmagic: sweet! |
| 16:04 |
JBoyer |
nfBurton, I'm building another master install real quick to see if I run into the same problem |
| 16:05 |
nfBurton |
Thanks |
| 16:14 |
JBoyer |
Well, for better or worse it's working fine for me, so I can't see what might be going on here. |
| 16:15 |
|
mmorgan joined #evergreen |
| 16:16 |
nfBurton |
Oh boy, okay. Thanks for checking. |
| 16:16 |
JBoyer |
What version of OpenSRF did you install (or already have in place)? |
| 16:17 |
nfBurton |
It even does that for the default template so it's a bit of a head scratcher. Maybe I just can it all and start fresh |
| 16:17 |
nfBurton |
3.2 |
| 16:17 |
|
khuckins joined #evergreen |
| 16:17 |
jeff |
oh, wait. this was running master with non-stock templates you were getting that error? |
| 16:18 |
JBoyer |
I will say the /api/ in the url looks odd to me, but I don't know if that means anything other than I haven't paid attention to that yet. |
| 16:18 |
nfBurton |
working master branch with default templates. I'm trying to create my non-stock but ran into that issue with last upgrade. |
| 16:18 |
jeff |
ah. |
| 16:19 |
nfBurton |
I may just trash openils and start fresh |
| 16:19 |
JBoyer |
mv /openils /openils.prev maybe? :) |
| 16:20 |
* JBoyer |
has forgotten about small things when doing a 'rm -rf /openils' in the past |
| 16:20 |
nfBurton |
yeah |
| 16:23 |
jeff |
presumably there was a failure in the call to open-ils.cstore.direct.money.open_billable_transaction_summary.search -- you might find that error wherever your opensrf logs lie |
| 16:24 |
|
jihpringle joined #evergreen |
| 16:24 |
jeff |
but ensuring that you have a working IDL in place and all Evergreen opensrf services running successfully and a clean autogen.sh run... those might be good first things to check. |
| 16:25 |
jeff |
but also that tt2 file not existing makes me wonder about your apache config. |
| 16:26 |
jeff |
@decide more sanity checks or better remote diagnosis tools |
| 16:26 |
pinesol |
jeff: That's a tough one... |
| 16:27 |
JBoyer |
sometimes the bot really does have the right answer, heh. |
| 16:33 |
Dyrcona |
nfBurton: What Linux distro and release are you installing on? |
| 16:35 |
nfBurton |
xenial 16.04 |
| 16:36 |
* jeff |
attempts to think of creative ways to tie a staff search to a staff user from potentially incomplete logs |
| 16:37 |
JBoyer |
incomplete how? |
| 16:38 |
jeff |
on the test box in question, i'm pretty sure that some things (at least "gateway" logs) are not being saved. |
| 16:38 |
jeff |
(an oversight, not intentional) |
| 16:39 |
Dyrcona |
nfBurton: Never mind. What I was thinking of only applies to Debian Stretch and Debian Buster. |
| 16:39 |
JBoyer |
Ah, I thought you meant something like rate-limited drops or very short retention periods, etc. So long as you have everything normally in osrfsys.* you've got a connection from username to authtoken. |
| 16:40 |
jeff |
yeah, I'm just not sure I've a link from "this search" to "this auth token" |
| 16:40 |
Dyrcona |
nfBurton: The symptoms would also be that osrf_control --diagnostic [-l] shows none of the C services running: cstore, pcrud, auth, auth_internal.... |
| 16:41 |
JBoyer |
Oh, right, it's not like the osrfsys files actually have the search terms. Oops. |
| 16:42 |
|
yboston joined #evergreen |
| 16:50 |
|
mmorgan left #evergreen |
| 16:52 |
* Dyrcona |
signs out for today. |
| 18:12 |
|
jonadab joined #evergreen |
| 19:39 |
|
jvwoolf joined #evergreen |
| 20:11 |
|
sandbergja joined #evergreen |
| 22:33 |
|
sandbergja joined #evergreen |
| 22:56 |
|
jvwoolf joined #evergreen |
| 23:01 |
pinesol |
News from qatests: Failed Installing AngularJS web client <http://testing.evergreen-ils.org/~live/test.28.html#2019-10-24T23:00:48,595715594-0400 -0> |
| 23:01 |
pinesol |
News from qatests: Failed Installing Angular web client <http://testing.evergreen-ils.org/~live/test.29.html#2019-10-24T23:00:48,642897535-0400 -2> |
| 23:01 |
pinesol |
News from qatests: Failed Log Output: osrfsys.log <http://testing.evergreen-ils.org/~live/test.76.html#2019-10-24T23:00:48,690042733-0400 -4> |
| 23:17 |
|
sandbergja joined #evergreen |
