| Time |
Nick |
Message |
| 05:45 |
|
book` joined #evergreen |
| 05:48 |
|
remingtron joined #evergreen |
| 05:48 |
|
StomproJ joined #evergreen |
| 05:55 |
|
remingtron joined #evergreen |
| 05:55 |
|
StomproJ joined #evergreen |
| 06:30 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 07:13 |
|
rjackson_isl joined #evergreen |
| 07:34 |
|
agoben joined #evergreen |
| 07:36 |
|
dwgreen joined #evergreen |
| 08:43 |
|
jvwoolf joined #evergreen |
| 08:48 |
JBoyer |
Hallo web team: "Unauthorized access to internal API. Please refer to https://support.google.com/customsearch/answer/4542055" |
| 08:48 |
JBoyer |
Google custom search is most unhappy on evergreen-ils.org/ |
| 08:48 |
JBoyer |
. |
| 08:53 |
rhamby |
JBoyer: bah, humbug |
| 08:54 |
|
jvwoolf1 joined #evergreen |
| 08:55 |
rhamby |
I actually want to rebuild and refresh the web site but I don't have the free cycles for it.... |
| 08:55 |
JBoyer |
rhamby, you will be visited this night by the 3 spirits, the ghost of APIs past, the ghost of APIs present, and the ghost of APIs future... |
| 08:56 |
rhamby |
JBoyer: can the ghost of API future learn how to document worth a #*$&? |
| 08:56 |
JBoyer |
It's just a /wsdl.xml in a hood. |
| 08:57 |
rhamby |
I'll send Timmy down to the grocers to get a stylesheet for two shillings on Christmas morn and call it a day then. |
| 08:59 |
|
bos20k joined #evergreen |
| 09:00 |
JBoyer |
foreach $one in @all { bless; } |
| 09:00 |
JBoyer |
Hmm, not even close. Humbug indees. |
| 09:01 |
JBoyer |
indeed, even. |
| 09:02 |
csharp |
JBoyer: sounds like a harrowing story |
| 09:10 |
remingtron |
webteam: Many pages on evergreen-ils.org are returning no content, just a tiny smily face |
| 09:10 |
remingtron |
For example: https://evergreen-ils.org/egdownloads/ |
| 09:11 |
rhamby |
let me look, I disabled the search plugin, may need to remove it |
| 09:19 |
dbwells |
rhamby: I just updated our library's Google custom search about a week or so ago. Do you want me to look at doing the same for evergreen-ils.org? I probably have all the access I need. |
| 09:19 |
rhamby |
dbwells: sure go for it |
| 09:20 |
dbwells |
rhamby: okay, I'll take a look |
| 09:22 |
dbwells |
rhamby: did you already remove the search plugin? |
| 09:23 |
abneiman |
remingtron: rhamby: https://evergreen-ils.org/communicate/calendar/ is returninig empty besides the header; nary a smiley face, either |
| 09:23 |
rhamby |
dbwells: I did and it didn't seem to have any effect |
| 09:24 |
remingtron |
abneiman: I see the smiley on that page too. You just have to look veeeery carefully. |
| 09:24 |
rhamby |
yeah, I still see the smiley too |
| 09:25 |
abneiman |
ah yes, almost falling off the left edge of my screen. well thank goodness for that! :) |
| 09:29 |
dbwells |
content should be back now in a quick and dirty fix, now on to the real fix |
| 09:30 |
rhamby |
dbwells++ |
| 09:31 |
rhamby |
have I ever mentioned that I have a kind of love/hate relationship with wordpress? |
| 09:35 |
|
yboston joined #evergreen |
| 09:54 |
abneiman |
dbwells++ |
| 10:07 |
|
Dyrcona joined #evergreen |
| 10:36 |
Dyrcona |
Whee! Due dates manually edited to about 2,000 years in the past. |
| 11:01 |
dbs |
Ah the ever-popular open-ils.storage drones going nuts for no very clear reason :( |
| 11:02 |
dbs |
"Collision using internal ID as tcn_value for record 3119620", I wonder if that's the underlying reason... |
| 11:25 |
dbs |
hah, ERROR: date/time field value out of range: "0000-12-29T23:59:00-04:00" |
| 11:30 |
dbs |
or is it crap like "open-ils.search[11996]: [ERR :11996:Z3950.pm:494:1513698318917678] z3950: bad XML : Tag "7||" is not a valid tag. at /usr/share/perl5/MARC/File/USMARC.pm line 223" |
| 11:31 |
|
Christineb joined #evergreen |
| 11:31 |
Dyrcona |
So, updated a test system to the latest ubuntu 16.04 4.4.0 kernel and other packages. Apache2 just times out. |
| 11:32 |
Dyrcona |
Don't seem to have this problem on a different machine. |
| 11:34 |
Dyrcona |
Ah, wait a minute.... |
| 11:36 |
dbs |
got it, hours of operation were set to 00:00:00 to 00:00:00 across the board |
| 11:38 |
Dyrcona |
Eh, mine's not that easy. ;) |
| 11:38 |
dbs |
This makes calculating due dates very difficult :) |
| 11:38 |
Dyrcona |
Yes, I think it would do that. |
| 11:42 |
pastebot |
"Dyrcona" at 64.57.241.14 pasted "Nifty...." (8 lines) at http://paste.evergreen-ils.org/958 |
| 11:43 |
Dyrcona |
Of course, the interface is UP RUNNING.... |
| 11:45 |
|
jihpringle joined #evergreen |
| 11:46 |
|
_adb joined #evergreen |
| 11:46 |
* Dyrcona |
thinks maybe kernel -104 has issues....But! my test VM doesn't do this. |
| 11:50 |
|
jvwoolf joined #evergreen |
| 11:52 |
jeff |
the "RTNETLINK answers: File exists" often indicates that state between your networking scripts and the live system has gotten out of sync. Sometimes it's your config trying to add a route that you've already added by hand, etc. Other times, a symptom of having two interfaces that both declare an identical route. |
| 11:53 |
jeff |
check your config for duplicate routes (including default routes / gateway entries), and then consider a full stop/start of networking or even a reboot. |
| 11:54 |
jeff |
also, make sure you're not fighting dhclient or similar, if the system originally had an interface configured for dhcp and you've changed it to static without stopping networking and/or dhclient. |
| 11:54 |
jeff |
(though that usually manifests in a different way) |
| 11:56 |
jeff |
(like, dhclient changing your "static" IP out from under you or removing IPs when the lease runs out and it can't contact a DHCP server to renew) |
| 11:59 |
Dyrcona |
jeff: I've done that. I just need to decide on the better way to fix it. |
| 12:33 |
jeff |
ah, just wanted to throw that out there if you hadn't considered yet -- to save you from chasing kernel issues needlessly. :-) |
| 12:33 |
Dyrcona |
Actually, I answered hastily because I wanted to get lunch. I am still looking into which of the many causes of that message I have. :) |
| 12:34 |
Dyrcona |
This only started after applying updates. I did NOT touch the network configuration in the meantime. (Someone else may have, but I doubt it.) I have also not manually brought networking or interfaces up or down. |
| 12:35 |
Dyrcona |
As usual, I'm doing more than 1 thing at once. :) |
| 12:37 |
|
khuckins joined #evergreen |
| 12:40 |
Dyrcona |
So many different possibilities.... |
| 12:46 |
Dyrcona |
So, it's looking like the problem is two gateway entries on different networks for different interfaces, but our training server has that (on an old kernel/release) and it works fine. |
| 12:46 |
Dyrcona |
But nothing I've tried so far has worked. |
| 12:49 |
dbs |
Current wishlist item: copy cataloguing via URI... just drop a URL that points at a MARCXML record into a form, and have the client go grab the record |
| 12:50 |
Dyrcona |
dbs: Or drop a file of records. |
| 12:50 |
dbs |
Dyrcona: yeah sure, for batch import |
| 12:50 |
Dyrcona |
So, yeah, looks like a routing issue..... But, why now? |
| 12:52 |
jeff |
Dyrcona: You have two different interfaces configured to start and both declare a gateway? |
| 12:53 |
Dyrcona |
Yes, on different networks. (Also, note: I did not set this up. At least, I don't think I did.) |
| 12:53 |
Dyrcona |
What I find strange is it worked, until yesterday. |
| 12:54 |
jeff |
Dyrcona: that's an unusual config and not something I'd expect to work without additional... cleverness. Perhaps the other system where this configuration exists has mitigating circumstances, such as one interface isn't set auto, or the init system (systemd or upstart or other) is slightly more tolerant of the error. |
| 12:54 |
Dyrcona |
jeff: Or someone manually messed with routes and it will break on reboot. :) |
| 12:55 |
jeff |
indeed. compare uptime vs mtime of /etc/network/interfaces or similar? :-) |
| 12:55 |
Dyrcona |
And, yeah, two gateways isn't supposed to work. |
| 12:55 |
Dyrcona |
Well, you can mess with routes in ways that don't touch interfaces. |
| 12:56 |
* jeff |
nods |
| 12:58 |
Dyrcona |
What makes this doubly funny is both gateways are different interfaces on the same "machine." |
| 13:01 |
Dyrcona |
When did route stop supporting flush? |
| 13:03 |
Dyrcona |
Dropping the gateway for the external address doesn't help. |
| 13:06 |
Dyrcona |
And, I really hate messing with networking remotely. |
| 13:07 |
Dyrcona |
Yeap. restart networking and lose my connection. |
| 13:09 |
Dyrcona |
Or rather not.... Just looked like I did but it came back. |
| 13:09 |
Dyrcona |
So, guess it is fixed for now. :) |
| 13:09 |
csharp |
sometimes fussing with /etc/udev/rules.d/* helps |
| 13:10 |
Dyrcona |
See, udev and systemd overly complicate things. I'm going to BSD and staying there. :) |
| 13:11 |
Dyrcona |
Apple/Mac OS X: Think different! |
| 13:11 |
Dyrcona |
BSD: Think correctly! |
| 13:26 |
JBoyer |
dbs: when you say you've got bonkers storage drones, do you mean they're pulling 100% on a core or dying or something altogether different? |
| 13:27 |
JBoyer |
I regularly see 1-6 just spinning full speed on select() |
| 13:59 |
Bmagic |
Anyone else running the new non-Ruby EDI pusher with B&T? |
| 14:01 |
JBoyer |
Bmagic, not yet. what kinds of issues are you seeing though? |
| 14:01 |
Bmagic |
I set it up with the defaults from Evergreen for B&T. They received our first order and complained about the placement of the SAN |
| 14:05 |
Bmagic |
Looking at the old format next to the new, it's very very close NAD+BY+{purchasersan vendorcode} instead of NAD+BY+{vendorsan vendorcode} (This is ORDERS) |
| 14:15 |
dbs |
JBoyer: yeah, 7-8 at 70% |
| 14:16 |
dbs |
Pretty sure the cause this time was the "actor.hours_of_operation says the library is never open but we're trying to calculate 21 open days OMG the infinity is beautiful......." |
| 14:19 |
jeff |
Dyrcona: as "luck" would have it I was just reminded of another common way to run into that madness with networking from earlier today: changing the IP of an interface in /etc/network/interfaces while the interface is still up can (possibly recently) result in that. |
| 14:19 |
Dyrcona |
jeff: Yes, in that case you have to ip addr flush dev [device] |
| 14:20 |
Dyrcona |
But, that was not my problem. Two gateways was my problem, and removing the gateway for the "LAN"IP fixed it. |
| 14:25 |
Dyrcona |
Is it just me or has networking gotten more brittle since the 3.x kernels? |
| 14:25 |
Dyrcona |
Maybe I should ask that in a Linux channel. :) |
| 14:29 |
berick |
Bmagic: looks like a bug.. that should def. be buyer SAN not vendor SAN in the NAD+BY |
| 14:29 |
berick |
Bmagic: looks like EDIWriter.pm line 172 |
| 14:30 |
berick |
should be org_unit_san instead of vendor_san |
| 14:30 |
Bmagic |
oh! |
| 14:31 |
Bmagic |
berick: you found the bug? |
| 14:31 |
berick |
Bmagic: yeah, untested, but i'm positive that's it |
| 14:32 |
pastebot |
"berick" at 64.57.241.14 pasted "for Bmagic" (13 lines) at http://paste.evergreen-ils.org/959 |
| 14:33 |
jeff |
@band add Untested Certainty |
| 14:33 |
pinesol_green |
jeff: Band 'Untested Certainty' added to list |
| 14:34 |
JBoyer |
dbs, I would hope that the only places that are closed 7 days a week have no copies, but just in case 7 places are about to be open for an hour a week... |
| 14:34 |
JBoyer |
dbs++ |
| 14:35 |
Bmagic |
berick++ |
| 14:35 |
Bmagic |
berick: I am going to test it! |
| 14:48 |
Dyrcona |
Bmagic++ #testing in production |
| 14:48 |
Bmagic |
yep! Exactly |
| 14:49 |
Bmagic |
hmmm... can I delete the acq.edi_message row? and the new order pusher will create it again? |
| 14:52 |
berick |
Bmagic: yes. you should also be able to set the status on the edi_message to 'retry' |
| 14:52 |
Bmagic |
according to the notes in the edi_order_pusher, I think I can |
| 14:53 |
Bmagic |
if I set it to retry, will it generate the new message? I think I am going to delete the row |
| 14:53 |
dbs |
JBoyer: yeah, I had automated our hours based on an external calendar--basically, look up next week's hours and apply them now so that people have a way of planning ahead |
| 14:53 |
berick |
yes, it generates a new value for the edi field |
| 14:53 |
dbs |
JBoyer: next week we're closed all week so that caused issues ;) |
| 14:54 |
dbs |
per http://git.evergreen-ils.org/?p=contrib/Conifer.git;a=commitdiff;h=c13ece121ebce1deaf47664b4c5e014b20bd0701 we now have a fallback |
| 14:57 |
Bmagic |
berick: great! The new code just executed and the order has been sent. Cross fingers |
| 14:57 |
berick |
Bmagic++ |
| 15:01 |
Bmagic |
Dyrcona++ # keeping insane presence on IRC |
| 15:08 |
jeffdavis |
dbs: would adding closed dates (rather than modifying hours of operation) prevent that issue? |
| 15:10 |
csharp |
I remember an issue where we had to have at least one day that was "open" for org units to behave correctly |
| 15:12 |
csharp |
from the 1.6/2011 days iirc |
| 15:18 |
dbs |
jeffdavis: it would avoid one problem, but we also want to accurately reflect our opening hours at the likes of https://laurentian.concat.ca/eg/opac/library/OSUL |
| 15:24 |
jeff |
wishlist: add support for special hours. :-) |
| 15:24 |
jeff |
(probably need some db and tpac changes, at least) |
| 16:03 |
|
khuckins_ joined #evergreen |
| 17:00 |
|
khuckins_ joined #evergreen |
| 17:04 |
|
khuckins__ joined #evergreen |
| 17:20 |
gmcharlt |
floating a proposal - given that there are some additional bugfixes I'd like to get into 3.0.3, I propose moving the maintenance release date to 12/289 |
| 17:20 |
gmcharlt |
er, 12/28 |
| 17:20 |
gmcharlt |
I'll be in a position to help write release notes next week and roll tarballs along with Bmagic |
| 17:47 |
berick |
gmcharlt: could bought a lot more time with 12/289 |
| 18:02 |
Dyrcona |
gmcharlt: +1 if you're looking for votes. |
| 18:02 |
Dyrcona |
A better release is better than an early release. |
| 18:03 |
* Dyrcona |
*poof* |
| 18:05 |
gmcharlt |
berick: my TARDIS is on the blink, alas |
| 18:10 |
berick |
ah, well, you'll fix it yesterday |
| 18:10 |
berick |
+1 to release |
| 18:18 |
dbwells |
Just noticed yesterday was our eight year Evergreen anniversary. Cheers to all, and here's to another eight :) |
| 18:19 |
berick |
dbwells: huzzah! |
| 18:32 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 20:42 |
|
StomproJ joined #evergreen |
| 20:56 |
|
jvwoolf joined #evergreen |
