| Time |
Nick |
Message |
| 03:35 |
|
Christineb joined #evergreen |
| 06:01 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 07:32 |
|
rjackson_isl_hom joined #evergreen |
| 07:48 |
|
collum joined #evergreen |
| 08:28 |
|
mmorgan joined #evergreen |
| 08:31 |
|
mantis1 joined #evergreen |
| 08:31 |
|
rfrasur joined #evergreen |
| 08:47 |
|
awitter joined #evergreen |
| 08:52 |
|
alynn26 joined #evergreen |
| 08:55 |
|
alynn26_away joined #evergreen |
| 08:57 |
|
alynn26_away joined #evergreen |
| 09:04 |
|
jvwoolf joined #evergreen |
| 09:16 |
|
nfBurton joined #evergreen |
| 09:16 |
|
nfBurton2 joined #evergreen |
| 09:27 |
|
Dyrcona joined #evergreen |
| 09:43 |
|
Keith_isl joined #evergreen |
| 09:57 |
Dyrcona |
So, I rebuilt a VM from scratch thinking it would fix this error: Bad arg length for Socket::pack_sockaddr_in, length is 0, should be 4 at /usr/lib/x86_64-linux-gnu/perl/5.26/Socket.pm line 157. |
| 09:57 |
Dyrcona |
But, no, it didn't. |
| 09:58 |
Dyrcona |
This VM is bascially identical to another one that works. I get that error when running autogen.sh. |
| 10:09 |
Dyrcona |
I think that may be the memcached configuration in opensrf.xml. I copied one from production and I forgot to change it. |
| 10:11 |
Dyrcona |
Bingo! We have a winner! |
| 10:12 |
|
rjackson_isl_hom joined #evergreen |
| 10:17 |
berick |
tried on a whim, pleased it worked: hightlight some json in vim then !jq and it will format it inline. neat. |
| 10:30 |
mmorgan |
berick++ |
| 10:32 |
csharp_ |
berick: highlight with "visual"? |
| 10:32 |
berick |
csharp_: yeah |
| 10:33 |
csharp_ |
holy crap |
| 10:33 |
csharp_ |
https://www.youtube.com/watch?v=9CS7j5I6aOc&ab_channel=GregSteel |
| 10:34 |
csharp_ |
I like me some IRC, but I occasionally want to be able to post gifs/vids and not just URLs :-) |
| 10:35 |
csharp_ |
@upgrade #evergreen |
| 10:35 |
pinesol |
csharp_: zarro boogs found |
| 10:39 |
jeff |
csharp_: how are things with that change in place? |
| 10:39 |
csharp_ |
jeff: excellently so far |
| 10:40 |
csharp_ |
I'll sign off after a day or two of watching, but looks like a win |
| 10:40 |
csharp_ |
that issue is probably responsible for system load and other issues we've seen in the past too |
| 10:43 |
berick |
belated jeff++ csharp_++ |
| 10:44 |
mmorgan |
jeff++ csharp_++ |
| 10:46 |
jeff |
csharp_: did you have many standing penalties with an org unit of 1 before the upgrade, or were most/all of those introduced by the new unified messages feature? |
| 10:48 |
csharp_ |
jeff: uh.... I don't know - pretty sure most were at "1" because we want them visible across the consortium |
| 11:10 |
Dyrcona |
Most of ours are at org_unit 1: 226,286 out of 251,542. |
| 11:16 |
jeff |
looks like most of ours are system-level, save for the invalid contact penalties. |
| 11:19 |
|
jvwoolf joined #evergreen |
| 11:37 |
csharp_ |
jvwoolf: see bug 1959461 - jeff developed a fix that has so far solved our open-ils.actor service issue |
| 11:37 |
pinesol |
Launchpad bug 1959461 in Evergreen "Fleshing org unit on standing penalties should omit org unit children" [Undecided,Confirmed] https://launchpad.net/bugs/1959461 - Assigned to Jeff Godin (jgodin) |
| 11:38 |
csharp_ |
it's a minor JS file change that's easily reverted (meaning that you could experiment like I have without too much trouble rolling back) |
| 11:59 |
csharp_ |
(also mantis1 ^^ - not sure who makes the call on code stuff at Biblio :-) ) |
| 12:06 |
jvwoolf |
jeff++ |
| 12:07 |
jvwoolf |
csharp_++ |
| 12:10 |
mantis1 |
csharp++ |
| 12:11 |
mantis1 |
jeff++ |
| 12:12 |
|
jihpringle joined #evergreen |
| 12:15 |
JBoyer |
jeff ++ |
| 12:15 |
JBoyer |
csharp_ ++ |
| 12:15 |
JBoyer |
Ooh, that's a neat trick, berick ++ |
| 12:20 |
csharp_ |
drone counts are lower across the board too |
| 12:21 |
jeff |
good to hear. |
| 12:21 |
csharp_ |
so maybe a good troubleshooting practice to do packet captures when facing anything like this? Jeff was suggesting that from the beginning\ |
| 12:22 |
jeff |
the patron update API call doesn't need several things that we currently send it. I think it's still a good idea to remove children on the org unit when locally fleshing it (the current branch in that bug), but I think there's also some things we should stop sending in the patron update call, and of course there's still the unresolved issue of "services shouldn't fall over in the face of this" |
| 12:25 |
Dyrcona |
jeff: I see something else (though not strictly related) in logs from a test VM that bug me along similar lines. Looks like 1 backend call leads to 200+ on our system when it could probably be a single JSON query. |
| 12:27 |
csharp_ |
jeff++ |
| 12:30 |
|
rjackson_isl_hom joined #evergreen |
| 12:30 |
jeff |
Dyrcona: yeah, that does sound bad. |
| 13:25 |
|
Guest61 joined #evergreen |
| 13:26 |
pastebot |
"Guest61" at 168.25.130.30 pasted "OpenPort" (4 lines) at http://paste.evergreen-ils.org/14446 |
| 13:27 |
* jeff |
responds to dteston's paste |
| 13:27 |
Guest61 |
Thank you jeff |
| 14:18 |
csharp_ |
okay, pastebot is dead for now |
| 14:18 |
csharp_ |
everyone use pastebin.com or similar :-) |
| 14:18 |
csharp_ |
it was infested with spam |
| 14:18 |
csharp_ |
and the project is super duper dead |
| 14:19 |
csharp_ |
like last update 2006 dead |
| 14:26 |
Dyrcona |
jeff++ |
| 14:27 |
Dyrcona |
csharp_++ |
| 14:30 |
Dyrcona |
Does anyone know how to debug issues with Chrome (or Firefox for that matter) when the problem is so bad that the developer tools freeze? |
| 14:33 |
csharp_ |
oof |
| 14:33 |
csharp_ |
I've seen that when IndexDB is struggling |
| 14:34 |
csharp_ |
Indexed DB rather |
| 14:35 |
csharp_ |
but that was back when we were trying to get standalone working |
| 14:35 |
Dyrcona |
Well, I found a Chromium debugger guide for Chromium 71, and I'm not sure I want to spend the afternoon with gdb. :) |
| 14:35 |
* Dyrcona |
is a bit rusty with gdb. |
| 14:36 |
Dyrcona |
Well, I get this even when I've cleared all local storage. I'll try again. |
| 14:39 |
Dyrcona |
Not much in IndexedDB either. Looks like an object for each org unit, a dozen or so settings, and that's about it. |
| 14:43 |
Dyrcona |
Part of me wants to put this on Launchpad, but it seems to only happen with our data. Using the same code on a concerto database, this doesn't happen. |
| 14:44 |
Dyrcona |
Why something in the database would cause Chrome to freeze like this is beyond me, and I can't figure out what call is doing it because everything from the logs seems to be all right when I run it individually. |
| 14:45 |
Dyrcona |
It happens with Firefox, too. |
| 14:48 |
Dyrcona |
So, in Firefox, I get this as that last console entry: Net: request open-ils.pcrud.search.pgtde.atomic |
| 14:49 |
Dyrcona |
That's also the last log entry in syslog, but that looks like it returns OK in the syslog. |
| 14:50 |
|
awitter joined #evergreen |
| 14:57 |
jeff |
can you reproduce on another machine, or in a fresh (Chrome) incognito window? |
| 14:58 |
jeff |
(ruling out any issues with the site data in the local browser profile) |
| 14:58 |
Dyrcona |
All right. I deleted all entries from permission.grp_tree_display_entry and it "works," except that I'm getting an error in the Firefox console with Evergreen 3.5. |
| 14:58 |
berick |
Dyrcona: a loop in your group tree? |
| 14:58 |
Dyrcona |
jeff: I've reproduced it multiple times, and another user reported it to me before I saw it. |
| 14:58 |
Dyrcona |
berick: Could be. |
| 15:00 |
Dyrcona |
I'll take a look at that in a db where I didn't delete them. :) |
| 15:01 |
Dyrcona |
That console error pops up no matter what I click on, even the hamburger menu. |
| 15:02 |
Dyrcona |
ERROR TypeError: l is undefined formatForDisplay https://jasontest.cwmars.org/eg2/en-US/4.b53e4ed52f806327df5b.js:1 |
| 15:02 |
Dyrcona |
set applyOrgId https://jasontest.cwmars.org/eg2/en-US/4.b53e4ed52f806327df5b.js:1 |
| 15:02 |
Dyrcona |
Ts https://jasontest.cwmars.org/eg2/en-US/main.22619f21d464835cbefc.js:2 |
| 15:02 |
Dyrcona |
No https://jasontest.cwmars.org/eg2/en-US/main.22619f21d464835cbefc.js:2 |
| 15:02 |
Dyrcona |
No https://jasontest.cwmars.org/eg2/en-US/main.22619f21d464835cbefc.js:2 |
| 15:02 |
Dyrcona |
No https://jasontest.cwmars.org/eg2/en-US/main.22619f21d464835cbefc.js:2 |
| 15:02 |
Dyrcona |
sa https://jasontest.cwmars.org/eg2/en-US/main.22619f21d464835cbefc.js:2 |
| 15:02 |
Dyrcona |
Ti https://jasontest.cwmars.org/eg2/en-US/14.628ba9007e32b90118df.js:1 |
| 15:02 |
Dyrcona |
updateDirectives https://jasontest.cwmars.org/eg2/en-US/main.22619f21d464835cbefc.js:2 |
| 15:02 |
Dyrcona |
Ro https://jasontest.cwmars.org/eg2/en-US/main.22619f21d464835cbefc.js:2 |
| 15:02 |
Dyrcona |
$o https://jasontest.cwmars.org/eg2/en-US/main.22619f21d464835cbefc.js:2 |
| 15:02 |
Dyrcona |
main.22619f21d464835cbefc.js:2:23796 |
| 15:02 |
jeff |
hey, we maintain a pastebin for a reason! <ducks> |
| 15:02 |
jeff |
too soon? :-) |
| 15:05 |
Dyrcona |
Heh. |
| 15:05 |
* Dyrcona |
paid for Pastebin pro to support something that I use. |
| 15:05 |
Dyrcona |
It's scary how little money some of these projects get. |
| 15:06 |
Dyrcona |
berick++ |
| 15:07 |
Dyrcona |
berick | jeff: A quick pair of select count(*)s and I think something could be very wrong with our pgtde. It has 1,036 entries and pgt has only 137. |
| 15:07 |
jeff |
hah |
| 15:09 |
jeff |
well, it's per-org. |
| 15:09 |
jeff |
up to per-org, at least. |
| 15:10 |
jeff |
so it could have as many as (pgt * aou) entries |
| 15:13 |
Dyrcona |
OK. I hadn't looked, yet. |
| 15:13 |
Dyrcona |
I'll dump pgt and pgtde and look for possible issues. |
| 15:19 |
Dyrcona |
Actually, I meant to just paste the first line of that error, but I guess FF wasn't cooperating. :) |
| 15:28 |
Dyrcona |
berick: Can you think of some code/query that would detect loops in either pgt or pgtde? |
| 15:28 |
Dyrcona |
Just eyeballing the data in a spreadsheet isn't very efficient. |
| 15:38 |
Dyrcona |
Seems that I'm not the only one having fun with permission groups today. |
| 16:54 |
|
jvwoolf left #evergreen |
| 17:06 |
|
mmorgan left #evergreen |
| 17:59 |
|
nfBurton joined #evergreen |
| 18:01 |
pinesol |
News from qatests: Failed Installing Angular web client <http://testing.evergreen-ils.org/~live//archive/2022-01/2022-01-31_16:00:03/test.29.html> |
| 21:59 |
|
nfBurton joined #evergreen |
