| Time |
Nick |
Message |
| 07:00 |
|
collum joined #evergreen |
| 08:35 |
|
mmorgan joined #evergreen |
| 10:04 |
|
collum joined #evergreen |
| 10:42 |
|
sandbergja joined #evergreen |
| 11:03 |
|
Christineb joined #evergreen |
| 12:10 |
|
Dyrcona joined #evergreen |
| 12:15 |
Dyrcona |
I suspect that authority_control_fields.pl can be run in parallel on batches of different bib records. I'm not so sure about authority_authority_linker.pl, but I might try it. |
| 12:25 |
Dyrcona |
It's running a long time for the authority_control_fields.pl, even with my hack for Lp 2119022. I think if I split it up into smaller batches, say 1,000 records each, and run 4 or so at once, it might finish more quickly. |
| 12:25 |
pinesol |
Launchpad bug 2119022 in Evergreen "Unexpected behavior of --start_id and --end_id with authority_control_fields.pl" [Undecided,New] https://launchpad.net/bugs/2119022 - Assigned to Jason Stephenson (jstephenson) |
| 12:26 |
Dyrcona |
I suppose that I should clean up that branch and post it on the bug. |
| 12:47 |
Dyrcona |
One person's bug is an other's feature. |
| 12:47 |
|
Lorne joined #evergreen |
| 12:52 |
jeff |
just remember to include an option to reenable spacebar heating, and you'll be all set. |
| 13:03 |
Dyrcona |
HCF ? |
| 13:26 |
|
Lorne joined #evergreen |
| 13:31 |
|
sandbergja joined #evergreen |
| 14:13 |
eeevil |
Dyrcona: we need to get the browse entry de-serialization branch in for authority_control_fields.pl to be parallelizable |
| 14:14 |
eeevil |
https://bugs.launchpad.net/evergreen/+bug/2091748 |
| 14:14 |
pinesol |
Launchpad bug 2091748 in Evergreen "Concurrent browse entry update/insert" [Wishlist,New] |
| 14:17 |
Dyrcona |
eeevil: Does it just end quietly? I'm getting no error output at all, which makes me think it isn't working. There's usually some messages about bad fields. |
| 14:19 |
* Dyrcona |
checks if that will apply to 3.7.4. :) |
| 14:20 |
eeevil |
what's "it" here? the authority linker? I don't ever run it personally so I don't know what its output looks like. but it causes ingests, and especially churn on the browse entry table, and that serializes everything |
| 14:20 |
Dyrcona |
authority_control_fields.pl. |
| 14:20 |
Dyrcona |
I didn't look super close, but though it could run in parallel, but guess not. |
| 14:21 |
Dyrcona |
s/though/thought/ |
| 14:21 |
eeevil |
I mean, it will run, but if you run 2 copies, A and B, it will end up serializing A-B-A-B-A-B |
| 14:22 |
eeevil |
because browse entry is a serialization point due to ingest locking. once that's no longer true (see: branch in that LP) then ALL INGESTS will be parallelizable |
| 14:22 |
Dyrcona |
Well, I also have the &> being called from exec in perl, which may or may not be going through the shell. |
| 14:22 |
eeevil |
this is why I've been asking for eyes on that for a year :) |
| 14:23 |
Dyrcona |
authority_control_fields.pl hits the browse entries? huh. |
| 14:24 |
eeevil |
that's the "insert $0s into bibs based on authority headings matches" script, right? |
| 14:24 |
Dyrcona |
And...... I don't have working as a remote on the vm..... another, huh. |
| 14:24 |
Dyrcona |
yes, that's it. |
| 14:25 |
Dyrcona |
It's not been my afternoon.... |
| 14:25 |
eeevil |
if it changes the bibs, it causes ingests |
| 14:25 |
Dyrcona |
OK. That makes sense. |
| 14:25 |
eeevil |
if records get ingested, the browse entry table gets touched |
| 14:30 |
Dyrcona |
oh man. This is a mess... I probably should not try it on something so old. :) |
| 14:31 |
eeevil |
yeah, it was not intended for backporting |
| 14:35 |
Dyrcona |
I might have a look at the diffs later. I can usually resolve the conflicts that way. |
| 14:36 |
Dyrcona |
I can try it on 3.15.3 in the meantime. |
| 15:00 |
Bmagic |
We found an undocumented feature in Evergreen: Server Administration -> User Setting Types -> column called "reg_default". When set, the patron registration will default check/uncheck the setting! |
| 15:00 |
Bmagic |
It answers the question I posted in IRC a couple of times: http://irc.evergreen-ils.org/evergreen/2025-04-22#i_577033 http://irc.evergreen-ils.org/evergreen/2025-04-17#i_576896 |
| 15:03 |
mmorgan |
Bmagic: I answered your question back on 4/22, but inadvertently latered the wrong nick :-( |
| 15:04 |
Bmagic |
mmorgan++ # OMG there it is |
| 15:04 |
Bmagic |
mmorgan: yes, that does the trick. To answer your question |
| 15:04 |
mmorgan |
:) |
| 15:06 |
* mmorgan |
looks for an LP bug about the missing documentation. |
| 15:13 |
eeevil |
that feature def could use some documentation! 6d77e5d64b180f5f78627547418757ecb2d7e4e5 is the commit in main that describes it |
| 15:17 |
mmorgan |
eeevil++ |
| 15:48 |
mmorgan |
bug 2119568 |
| 15:48 |
pinesol |
Launchpad bug 2119568 in Evergreen "Docs: User Setting Types and Registration Default Option" [Undecided,New] https://launchpad.net/bugs/2119568 |
| 16:16 |
Bmagic |
mmorgan++ |
| 17:01 |
|
mmorgan left #evergreen |
| 17:40 |
gmcharlt |
3.16 release schedule posted: https://wiki.evergreen-ils.org/doku.php?id=faqs:evergreen_roadmap |
| 17:40 |
gmcharlt |
bottom line: general release on 11 November |
| 17:44 |
jeff |
06 November? |
| 17:45 |
gmcharlt |
bah |
| 17:45 |
gmcharlt |
6 November - we just making a release, not ending a world war |
| 18:13 |
jeffdavis |
I feel like that would be a valid reason to bump the version number to 4.0. |
