| Time |
Nick |
Message |
| 07:52 |
|
BDorsey joined #evergreen |
| 08:26 |
|
mantis1 joined #evergreen |
| 08:33 |
|
dguarrac joined #evergreen |
| 08:33 |
|
kworstell-isl joined #evergreen |
| 08:35 |
|
jvwoolf joined #evergreen |
| 08:35 |
|
mmorgan joined #evergreen |
| 08:56 |
|
Dyrcona joined #evergreen |
| 09:00 |
|
BDorsey joined #evergreen |
| 09:39 |
|
Dyrcona joined #evergreen |
| 09:39 |
abneiman |
docs branch here, if I did things right --> https://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/abneiman/docs-releasenotes-2022-10-18 |
| 09:40 |
abneiman |
jihpringle++ jweston++ for their work & assistance! |
| 09:40 |
miker |
Dyrcona: thanks for throwing QI at a "real" test task |
| 09:40 |
mmorgan |
abneiman++ |
| 09:41 |
mmorgan |
jihpringle++ |
| 09:41 |
mmorgan |
jweston++ |
| 09:43 |
Dyrcona |
miker: Don't be so quick to thank me. Looks like ingest_ctl died on me. |
| 09:44 |
miker |
well, let's figure out the bugs /before/ the commit! :) |
| 09:46 |
Dyrcona |
I'm going to add the log to the bug. I'm suspicious of the message though. It says the database server closed the connection, but I have a psql that was sitting idle overnight that works just fine still. |
| 09:54 |
Dyrcona |
I'll start the coordinator over again and see what happens this time. |
| 09:57 |
Dyrcona |
So, I accidentally ran it with --status as an option, and it said that --status was an invalid option, but then I started getting errors on the console about updating some records. I think that's a bug, too. If an invalid cli option is supplied the program should probably exit immediately. |
| 10:01 |
Dyrcona |
And, bug updated with that information. |
| 10:18 |
miker |
Dyrcona: I don't see the log on the LP bug... |
| 10:19 |
miker |
but, yes, invalid options should cause it to stop right away. /me will ask the getopt::long docs how to do that |
| 10:34 |
miker |
well, that's just simple, isn't it. Dyrcona, I'll push an "or help()" patch, but that's all it takes. |
| 10:45 |
Dyrcona |
I meant to add it as an attachment. I guess I hit the wrong button/it didn't make it. |
| 10:46 |
Dyrcona |
Oh right. Chrome doesn't work with gvfs and sftp-mounted locations any more. |
| 10:49 |
miker |
Dyrcona: log suggests that there was an explosion in the DB causing a forced restart/recovery |
| 10:50 |
Dyrcona |
It seems to be running just fine, now Almost 9,000 records processed. |
| 10:50 |
Dyrcona |
Well, I'll check the postgres logs, but the Evergreen processes look OK this morning. |
| 10:54 |
miker |
hrm... if it's a transient issue, something we /should/ retry on (and, for the specific thing it was doing when errored out, it /is/) we probably want $raise_db_error to default to false |
| 11:03 |
Dyrcona |
Nothing in the logs about recovery mode, and grep doesn't turn up any relevant errors. |
| 11:03 |
Dyrcona |
The last error is me making a typo: 2022-10-17 16:41:35.864 EDT [83844] evergreen jasontest ERROR: column "enable" of relation "internal_flag" does not exist at character 33 |
| 11:06 |
|
Christineb joined #evergreen |
| 11:10 |
Dyrcona |
Timestamps in the queued ingest log would be helpful. |
| 11:18 |
miker |
I wonder if "LIMIT NULL" isn't allowed in newer PGs... |
| 11:18 |
|
jihpringle joined #evergreen |
| 11:18 |
Dyrcona |
Dunno. This is PG10. If that were the problem, I think it would just die. |
| 11:19 |
Dyrcona |
The last log entries from yesterday don't show anything special: 2022-10-17 17:20:14.129 EDT [84880] evergreenjasontest LOG: duration: 38334.23 |
| 11:19 |
Dyrcona |
4 ms execute <unnamed>: SELECT action.process_ingest_queue_entry($1) |
| 11:19 |
Dyrcona |
2022-10-17 17:20:14.129 EDT [84880] evergreenjasontest DETAIL: parameters: $1 |
| 11:19 |
Dyrcona |
= '4046' |
| 11:19 |
Dyrcona |
After that one, the process_ingest_queue_entry lines just stop. No error; not a hint as to why. |
| 11:20 |
Dyrcona |
The biblio.record_entry updates from my other process continue. |
| 11:21 |
miker |
the log first dies on a select gathering entries in sub gather_outstanding_nonqueue_entries(), which calls refresh_main_db to make sure it's connected before the query starts ... but maybe that's a red herring |
| 11:23 |
Dyrcona |
It's still chugging away today, but I have nothing else going on right now. |
| 11:52 |
Dyrcona |
I've been looking for other errors in the o/s logs on both the virtual machine and the db server. I haven't found any. |
| 13:19 |
|
rfrasur joined #evergreen |
| 13:40 |
rhamby |
the hack-a-way signup is now live for next week - https://hopin.com/events/evergreen-hack-a-way-2022 |
| 13:45 |
Dyrcona |
rhamby++ |
| 13:57 |
|
BDorsey_ joined #evergreen |
| 13:57 |
|
rfrasur_ joined #evergreen |
| 14:34 |
|
kworstell-isl joined #evergreen |
| 14:49 |
Bmagic |
rhamby++ |
| 14:52 |
csharp_ |
rhamby++ |
| 15:25 |
jeffdavis |
csharp_: have you seen bug 1990306 ? I think you were seeing something similar on your system back in Feb, judging from my IRC logs |
| 15:25 |
pinesol |
Launchpad bug 1990306 in Evergreen "Flood of open-ils.actor.user.has_work_perm_at.batch requests for VIEW_USER with null authtoken" [Undecided,New] https://launchpad.net/bugs/1990306 |
| 15:27 |
csharp_ |
jeffdavis: yes, seen it pretty recently too |
| 15:38 |
|
mmorgan left #evergreen |
| 15:45 |
|
jvwoolf left #evergreen |
| 15:52 |
|
jihpringle joined #evergreen |
| 16:01 |
|
mmorgan joined #evergreen |
| 16:55 |
mmorgan |
claiming 1339 |
| 17:27 |
pinesol |
News from commits: LP1913604 - Stamping upgrade script <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=2d4bbf394ba44c057210c7fa1759eea88116cb2b> |
| 17:27 |
pinesol |
News from commits: LP1913604: Course materials module should remember original circ_lib <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=2479659ffd59cbed59afdbd2b723c74e339ec7e5> |
| 17:27 |
pinesol |
News from commits: LP1913604: course materials module shouldn't move items to different owning libs <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=8bfad095ec8cd16e44f568ca5cc61d060672b6c8> |
| 17:27 |
pinesol |
News from commits: LP1913604: Alert staff when associating item with course at a different library <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=654a51bd22b003653a475ad2e3790ea404fd7b53> |
| 17:27 |
pinesol |
News from commits: LP1940105: Perl detach material correctly resets item location <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=c4bfc19e4035278de572ba6c044a97bb8f9bcf32> |
| 17:27 |
pinesol |
News from commits: lp1940105 Archive Course Should Use detach_material <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=5e812fb64f612efd5cecb18a97a4c857dc4da3d7> |
| 17:30 |
|
mmorgan left #evergreen |
| 23:49 |
|
book` joined #evergreen |
