| Time |
Nick |
Message |
| 07:08 |
|
collum joined #evergreen |
| 07:59 |
|
BDorsey joined #evergreen |
| 08:22 |
|
Dyrcona joined #evergreen |
| 08:28 |
|
mantis joined #evergreen |
| 08:33 |
|
redavis joined #evergreen |
| 08:36 |
|
mmorgan joined #evergreen |
| 09:06 |
|
dguarrac joined #evergreen |
| 09:09 |
Dyrcona |
Puzzle of the day: Why does open-ils.cat.biblio.record.metadata.retrieve check for the VIEW_USER permission? |
| 09:16 |
Dyrcona |
Related question: I have a metarecord ID. I want to get the master bib record or list of bib records associated with the metarecord. What's the best way? open-ils.pcrud? |
| 09:19 |
Dyrcona |
Well, there is open-ils.supercat.metarecord.mods.retrieve, which might do the trick almost. |
| 09:19 |
Dyrcona |
In this case, I may not need the bib ID. The mods might be enough.... not sure. |
| 09:24 |
mmorgan |
Dyrcona: re: open-ils.cat.biblio.record.metadata.retrieve, suspect it's to access the creator and editor. |
| 09:29 |
berick |
Dyrcona: pcrud to metabib.metarecord and metabib.metarecord_source_map will do it. |
| 09:35 |
Dyrcona |
mmorgan++ berick++ |
| 09:36 |
Dyrcona |
I want to get the metarecord info for Aspen, so I'm going to see if the mods is enough. I think something like that goes on with copy holds, but I'm now in the middle of updating my test/dev vms. |
| 09:39 |
Dyrcona |
Hm.. looks like my db server is still booting or failed to boot all the way... |
| 09:39 |
|
sandbergja joined #evergreen |
| 09:42 |
Dyrcona |
Yeah. Power was off. I'm pretty sure I did 'reboot' and not 'shutdown' |
| 09:44 |
Dyrcona |
Maybe I should consider early retirement? I was looking at the old server, not the new one.... |
| 09:51 |
Bmagic_ |
lol |
| 09:52 |
Dyrcona |
Looks like the supercat function also returns the master record id, so I've got options. |
| 09:57 |
* Dyrcona |
doesn't use pcrud enough to have the "formulas" memorized. |
| 09:57 |
redavis |
csharp_, pinging you to determine your availability today to upload "the things" to the server today (or tomorrow if needs be). |
| 09:58 |
sandbergja |
mantis++ |
| 09:59 |
Dyrcona |
redavis: If csharp_ is not available, I am. |
| 09:59 |
redavis |
Dyrcona++ |
| 10:01 |
Dyrcona |
Helps to have an authtoken for pcrud. Looks like the dummy authtoken would work for this request, too, but since the target system will be logged in, might as well use the actual authtoken. |
| 10:03 |
redavis |
mantis++ |
| 10:25 |
sandbergja |
mmorgan++ |
| 10:26 |
redavis |
mmorgan++ |
| 10:54 |
Dyrcona |
pinesol: Can decide handle more than 1 option? |
| 10:54 |
pinesol |
Dyrcona: https://photos.app.goo.gl/jhfeaL3dLJaVRjTr5 |
| 10:55 |
Dyrcona |
@help decide |
| 10:55 |
pinesol |
Dyrcona: (decide ) -- |
| 10:55 |
Dyrcona |
csharp_ Looking so good! |
| 10:56 |
Dyrcona |
@decide Jersey Mike's or Wootrucks or Chick-fil-a |
| 10:56 |
pinesol |
Dyrcona: go with Jersey Mike's |
| 10:58 |
Bmagic |
csharp_++ |
| 11:14 |
redavis |
That's pretty amazing. csharp_++ |
| 11:15 |
Dyrcona |
Huh. Router crashed with Redis. Not sure why. |
| 11:17 |
Dyrcona |
service router pid=934948 is running with no pidfile! use --force-clean-process to automatically kill orphan processes # From restart-all.... |
| 11:17 |
Dyrcona |
osrf_control seems hung and not responding to SIGINT. |
| 11:25 |
Dyrcona |
I suspect that ResiSRF is not there, yet, but maybe by the fall. |
| 11:26 |
berick |
Dyrcona: segfault? if you can recreate / get backtrace I'm all ears. |
| 11:43 |
Dyrcona |
I'll see if I can find anything in the logs. |
| 11:43 |
Dyrcona |
I ended up install Ubuntu updates and restarting it. |
| 12:08 |
Dyrcona |
pinesol: Good call on the Jersey Mike's. |
| 12:08 |
pinesol |
Dyrcona: GitHub is my copilot |
| 12:09 |
Dyrcona |
Ugh. I just got an email from GitHub saying that I have free access to Copilot. |
| 12:14 |
JBoyer |
"Writing to a scripting language with a JIT compiler on top of a remote document viewer not abstract enough for you? Just wait until you hear what we plan to do with Three Mile Island!" |
| 12:14 |
sleary |
>.< |
| 12:17 |
|
Christineb joined #evergreen |
| 12:20 |
Dyrcona |
JBoyer++ |
| 12:28 |
Dyrcona |
If I want to do a fleshed retrieve via pcrud, does it matter where I add "_fleshed?" |
| 12:32 |
Dyrcona |
Hrm. Apparently, fleshed PCRUD methods are not automagically created? |
| 12:36 |
Dyrcona |
Maybe it's my version of Evergreen? |
| 12:37 |
Dyrcona |
No, that's not it. I have the same comment about support fleshing with PCRUD. |
| 12:38 |
JBoyer |
I don't have an example handy but I didn't think _fleshed was an api name thing, but went somewhere in the options because it's assigned a depth and isn't just a boolean. |
| 12:38 |
JBoyer |
{blah: blah, fleshed: 2, blahblah: blerb}, etc. |
| 12:42 |
berick |
open-ils.pcrud.retrieve.au 1 {"flesh":1,"flesh_fields":{"au":["home_ou"]}} |
| 12:42 |
Dyrcona |
berick: I was using flesh_depth:1. I'll change it and see what happenes. |
| 12:45 |
Dyrcona |
And..... The field I'm trying flesh isn't a link after all..... |
| 12:45 |
Dyrcona |
Or, it's a link from bre, but I'm retrieving mmr. |
| 12:47 |
Dyrcona |
Never mind. The mods field isn't linked to anything. I will try fleshing the master_record to make sure I have the syntax correct. |
| 12:47 |
Dyrcona |
berick++ that worked! |
| 12:48 |
Dyrcona |
JBoyer: I think you're correct when it comes to pcrud and cstore, now that I've played with pcrud a bit. I don't usually try this from srfsh. |
| 12:53 |
Dyrcona |
This particular record's mods field is null anyway. |
| 12:55 |
Dyrcona |
I wonder if we missed an update somewhere? My test database has 1,623,928 metarecord entries with null mods, and 591 where it isn't null. |
| 12:58 |
Dyrcona |
Production has similar numbers. |
| 13:00 |
|
BDorsey_ joined #evergreen |
| 13:21 |
|
jvwoolf joined #evergreen |
| 13:59 |
csharp_ |
redavis: I'm here - sorry, was in meetings all morning and just back from lunch |
| 14:00 |
redavis |
No worries at all. csharp_, we're just waiting on 3.13.7 and testing tarballs |
| 14:00 |
csharp_ |
onky donky |
| 14:22 |
|
kworstell-isl joined #evergreen |
| 14:32 |
|
BDorsey__ joined #evergreen |
| 14:46 |
|
kworstell_isl joined #evergreen |
| 15:19 |
Dyrcona |
berick: lots of segfaults. I'll see if I can find any stack traces. journalctl reports no entries for opensrf_router. |
| 15:25 |
Dyrcona |
berick: No stack traces, sorry. |
| 15:27 |
Bmagic |
I was having segfaults during executions of action_trigger_runner on top of Redis, but only sometimes. And only in a docker container. |
| 15:27 |
Dyrcona |
Best I can do is say that a segfault occurred in libopnesrf.so.2.0.0 from code running in opensrf_router at 3:30 am. |
| 15:27 |
Dyrcona |
Bmagic: This was on dev.cwmars.org. |
| 15:28 |
Bmagic |
right, I assumed |
| 15:28 |
Bmagic |
what I'm saying is: that's interesting |
| 15:28 |
Dyrcona |
It was probably action trigger related, but don't run them all on there. |
| 15:29 |
Bmagic |
my current theory is size. A certain size of action trigger drone/thread is causing it. |
| 15:29 |
Dyrcona |
There were a bunch more at 10:30 am EST. |
| 15:30 |
berick |
assuming you'd have to run in gdb to get a trace |
| 15:30 |
berick |
run + reproduce |
| 15:31 |
Dyrcona |
yeah, probably would have to use gdb. I have seen useful stack info in journalctl for other things, though. That's why I looked. |
| 15:31 |
berick |
ah |
| 15:32 |
|
mantis left #evergreen |
| 15:38 |
Dyrcona |
Likely Autorenew blew it up. |
| 15:38 |
Dyrcona |
That starts at 2:30 am. |
| 15:47 |
Dyrcona |
I have a way to test that hypothesis. I have another machine with half the ram set up to use Redis. It has the same dump loaded. If I run autorenew right now, it should get almost the same data to process. |
| 15:48 |
Bmagic |
Dyrcona++ |
| 15:51 |
Dyrcona |
And, they're off! |
| 15:51 |
redavis |
sandbergja++ |
| 15:53 |
Dyrcona |
Probably won't find out until tomorrow morning. |
| 16:17 |
Dyrcona |
It hasn't crashed, yet, and I'm signing out. I'll check on it in the morning and report whatever happens. |
| 16:53 |
|
jvwoolf left #evergreen |
| 17:09 |
|
mmorgan left #evergreen |
