Evergreen ILS Website

IRC log for #evergreen, 2021-10-12

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat

All times shown according to the server's local time.

Time Nick Message
02:56 JBoyer joined #evergreen
02:56 jweston joined #evergreen
02:56 miker joined #evergreen
02:56 akilsdonk joined #evergreen
06:00 pinesol News from qatests: Testing Success <http://testing.evergreen-ils.org/~live>
06:38 abneiman joined #evergreen
07:22 rjackson_isl_hom joined #evergreen
08:02 Dyrcona joined #evergreen
08:28 mantis joined #evergreen
09:00 rfrasur joined #evergreen
09:14 collum joined #evergreen
10:34 jvwoolf joined #evergreen
12:13 Dyrcona So, from what it looks like (i.e. all of my tests say that the replication is caught up with zero lag), and from what RhodiumToad implied over in #postgresql, that starting replication query I mentioned yesterday is apparently normal in Pg 10.
12:15 jeff not sure I saw you mention that yesterday.
12:15 jeff (in here, anyway)
12:22 jihpringle joined #evergreen
12:26 Dyrcona jeff: You are correct. I did not mention it in here yesterday.
12:26 Dyrcona Anyway, there you go if you're using replication and planning to upgrade. It looks like the replication process show up in pg_stat_activity in Pg 10+.
12:27 Dyrcona On the primary database cluster, anyway.
12:28 Dyrcona Also, the change that I made to the routing on the load balancer seems to interfere with the VPN configuration we use. I'll have to make an adjustment, but I'll wait until later.
12:28 Dyrcona I dd mention that yesterday. :)
12:54 jeff ah, yep. from the PostgreSQL 10.0 changelog:
12:54 jeff Add pg_stat_activity reporting of low-level wait states (Michael Paquier, Robert Haas, Rushabh Lathia)
12:54 jeff This change enables reporting of numerous low-level wait conditions, including latch waits, file reads/writes/fsyncs, client reads/writes, and synchronous replication.
12:55 jeff looks like you can probably filter them out by backend_type (also new)
13:07 jeff joined #evergreen
13:13 Dyrcona I missed the significance of that.
13:13 Dyrcona I'll have to review the change log again.
13:13 Dyrcona jeff++
13:14 jeff (then some other changes in at least 10.2 and 10.18 related to walsenders and pg_stat_activity)
13:14 Dyrcona For now, we've modified our monitor query to ignore START_REPLICATION.
13:14 Dyrcona I think we're on 10.18.
13:14 * Dyrcona checks.
13:14 jeff 10.2 says: Show walsenders that are sending base backups as active in the pg_stat_activity view (Magnus Hagander)
13:15 Dyrcona Yeahp.
13:15 jeff and 10.18 says: Make walsenders show their latest replication commands in pg_stat_activity (Tom Lane) / Previously, a walsender would show its latest SQL command, which was confusing if it's now doing some replication operation instead. Now we show replication-protocol commands on the same footing as SQL commands.
13:15 Dyrcona Yeah, that's likely what we're seeing.
13:16 Dyrcona If it's just 1, I should probably check my walsenders configuration.
13:20 jeff that last bit (in 10.18) is quite recent (which makes sense, 10.18 itself being rather recent): https://www.postgresql.org/message-​id/880181.1600026471@sss.pgh.pa.us
13:21 Dyrcona Yeah. Also, I think my wal senders config is OK. It's all caught up after all, so it doesn't need to run more than 1. I have the max at 5.
13:22 jeff I don't think that you'd have more than one even if it was "catching up". It's one per WAL client/receiver, and not something that goes up and down depending on how much WAL traffic you're generating.
13:24 jeff depending on if you're using pg_basebackup with --wal-method=stream, you'd have an extra one running during (part of?) your backup.
13:25 jeff (or -Xs or --xlog-method, or however you or your version of postgres spells it) :-)
13:25 Dyrcona Makes sense. I don't usually pay much attention when pg_basebackup is running, certainly not yesterday. It took about 45 minutes to an hour.
13:59 pinesol [evergreen|Bill Erickson] LP1903358 Staff catalog holds barcode realtime lookup - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=7f38b6f>
14:17 pinesol [evergreen|Galen Charlton] i18n: fix contruct in seed data that broken string extraction - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=72a3811>
14:17 pinesol [evergreen|Galen Charlton] Translation updates - newpot - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=8687ba8>
14:31 JBoyer gmcharlt++
14:31 JBoyer Dev meeting in ~30
14:53 shulabear joined #evergreen
14:54 gmcharlt 3.8-beta is now available https://evergreen-ils.org/ev​ergreen-3-8-beta-available/
14:54 csharp_ gmcharlt++
14:57 JBoyer gmcharlt++
14:57 shulabear gmcharlt++
14:57 terranm joined #evergreen
15:02 JBoyer I suppose we may as well have a shindig since I did sent out an invitation.
15:02 JBoyer #startmeeting 2021-10-12 - Developer Meeting, Agenda Available at https://wiki.evergreen-ils.org/do​ku.php?id=dev:meetings:2021-10-12
15:02 pinesol Meeting started Tue Oct 12 15:02:33 2021 US/Eastern.  The chair is JBoyer. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:02 pinesol Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:02 pinesol The meeting name has been set to '2021_10_12___developer_meeting__agenda_​available_at_https___wiki_evergreen_ils_​org_doku_php_id_dev_meetings_2021_10_12'
15:02 JBoyer #topic Introductions
15:02 JBoyer Hello! have we met?
15:02 JBoyer #info JBoyer = Jason Boyer, EOLI
15:02 terranm #info terranm = Terran McCanna, PINES
15:03 phasefx #info phasefx = Jason Etheridge, EOLI
15:03 rfrasur #info rfrasur = Ruth Frasur, EI/ECDI
15:03 abneiman I don't know any of you weirdos
15:03 shulabear #info shulabear = Shula Link, PINES/GCHRL
15:03 Dyrcona #info Dyrcona = Jason Stephenson, CW MARS
15:03 gmcharlt #info gmcharlt = Galen Charlton, EOLI
15:03 abneiman #info abneiman = Andrea Buntz Neiman, Equinox
15:03 JBoyer abneiman++
15:04 JBoyer Folks joining later are welcome to #info up upon said arrivals.
15:05 JBoyer It's time for everyone's favorite game:
15:05 JBoyer #topic Action Items from Last Meeting
15:05 JBoyer #info JBoyer will exercise the staff client in a current release of Edge
15:05 terranm It's fun when they're not my action items
15:05 sandbergja joined #evergreen
15:05 JBoyer #info JBoyer will look at adding Edge on Windows support to the Hatch installer
15:05 JBoyer To be honest with myself as much as the rest of you, this just isn't happening until November at the earliest, and I don't have the highest hopes for that either.
15:05 JBoyer I can promise you that it will be an uneventful thing to test if anyone else would like to volunteer to experiment with the staff client in Edge.
15:06 sandbergja joined #evergreen
15:06 JBoyer Are there any volunteers interested in downloading Microsoft Chrome and making sure it Does The Thing?
15:06 sandbergja #info sandbergja = Jane Sandberg, Linn-Benton Community College, coming in late
15:07 shulabear I can poke at it some in Edge.
15:07 JBoyer shulabear++
15:09 JBoyer I don't expect any surprises since I do use it at home to work with Evergreen, but I use a pretty limited subset of the client. If you could go through and poke at most interfaces side by side with Chrome and note any differences that would go a long way towards answering the question of how hard it would be to support.
15:10 shulabear Got it.
15:10 JBoyer #action shulabear will test out the staff client in Microsoft Edge
15:10 terranm Speaking of which, has anybody done any extensive testing with ChromeOS? I know there are lots of printing headaches with ChromeOS, but not sure about any other potential problems
15:10 terranm89 joined #evergreen
15:11 JBoyer I mean, it literally is Chrome, even moreso than Edge, so I would hope it works, printing (and Hatch) aside.
15:11 terranm89 Love it when I ask a question and IRC immediately kicks me off
15:11 JBoyer Gotta turn down my Shatner filter
15:11 csharp_ # info csharp = Chris Sharp
15:11 csharp_ er...
15:11 csharp_ #info csharp = Chris Sharp
15:11 csharp_ there we go
15:12 JBoyer terranm89, all you missed was me making a word salad: "I mean, it literally is Chrome, even moreso than Edge, so I would hope it works, printing (and Hatch) aside."
15:12 terranm89 JBoyer++
15:12 * csharp_ hears everything JBoyer types in Shatnervoice now
15:12 JBoyer Just wait until my album comes out.
15:12 csharp_ JBoyer++
15:13 JBoyer As for the Hatch installer, I'm kind of on the fence about just removing all of the auto-install code and having links to the store pages in the client to install the extensions; anyone have thoughts on that?
15:13 shulabear I'm hearing Walken more than Shatner.
15:13 rfrasur shulabear - same.  For whatever reason.
15:13 csharp_ JBoyer: makes sense to me
15:13 terranm89 Few people know that JBoyer sings like Shatner AND dances like Walken
15:14 JBoyer More people know it than may want to, hey-o!
15:14 Dyrcona Walken can actually dance, so...
15:14 rfrasur omg, the mental images. And so much cowbell.
15:15 JBoyer My thinking with the Hatch links would be that they're only visible if the extension is missing and only the link you need to avoid confusion. If that was in place I'd be happy to have the registry shenanigans removed from the installer so long as berick doesn't object.
15:15 Dyrcona For later: https://www.youtube.com/watch?v=wCDIYvFmgW8
15:16 terranm89 I was just looking for that link
15:16 Dyrcona Well, I don't like registry shenanigans either.
15:16 JBoyer I think I know what it is, and will verify later.
15:17 JBoyer If there are no more tangentially Edge related thoughts we can move on to the exciting release update.
15:17 JBoyer #topic Evergreen Release Updates
15:17 Dyrcona gmcharlt++
15:17 JBoyer New beta just dropped, gmcharlt++
15:17 gmcharlt #info Evergreen 3.8-beta is avaialble
15:17 terranm89 gmcharlt++ releaseteam++
15:17 rfrasur gmcharlt++
15:18 gmcharlt some comments:
15:18 abneiman gmcharlt++
15:18 gmcharlt 1. we should also do an OpenSRF release, but that would mostly be about noting the Debian support changes
15:19 gmcharlt 2. For the release notes, I used 'asciidoctor' rather than 'asciidoc' to generate the HTML version. Main reason: asciidoctor supports the number of headings levels we're currently using, asciidoc does not
15:19 gmcharlt 3. testers welcome and needed
15:19 gmcharlt so I'm wondering if anybody can make commitments to test this week or early next week
15:20 gmcharlt assuming successful testing, I think the RC period could be truncated or potentially even dropped
15:21 JBoyer Note if anyone is on the fence that some of the bugsquash servers could be updated to have the 3.8 beta if you're concerned about not being able to install it locally.
15:22 rfrasur Are we just testing what's in the release notes?
15:22 rfrasur (sorry, I know this is a dumb question)
15:22 abneiman I can test against an updated bugsquash (need to finalize some docs things anyway)
15:22 JBoyer Making sure that nothing existing was broken would also be good.
15:22 gmcharlt rfrasur: that would be a started point, but the broader testing that we collectively do, the more regressions that get identified
15:22 rfrasur So, release notes and generalized "everything" testing?
15:23 gmcharlt yeah
15:23 JBoyer I'll just plan to load up pattypan and festivus with 3.8b. pattypan will be a fresh db, festivus an upgrade from 3.7.1
15:23 rfrasur If there's a server, I can commit some time.
15:23 JBoyer Those 2 will be done today
15:23 terranm89 I can't promise anything, but if it's on one of the bugsquash servers then I can try to poke at it between other things
15:25 rfrasur JBoyer, whenever they're ready... I'll have a little more availability on the weekend though.
15:25 JBoyer later today basically means plan to look at them whenever you have time tomorrow or later; I'm not working on them this particular moment. ;)
15:25 rfrasur lol, slacker
15:26 JBoyer Only one keyboard on this machine.
15:26 JBoyer So, if no one has anything else to add re: testing 3.8b we have our LP updates coming up.
15:27 terranm89 Do these LP tag changes sound okay to everyone? Any red flags or other tag suggestions? https://docs.google.com/document/d/1rB11V4​Lla_c5Sw_CuytvQRbDBaa6RsIgGA6ZswaaUhs/edit#
15:27 terranm89 Sorry, did I jump ahead?
15:27 JBoyer Technically, but that's LP too. :)
15:27 terranm89 I saw LP and panicked
15:28 JBoyer I'll just copy / paste a few lines while people take the time to look that over.
15:28 JBoyer #topic Launchpad Status (as of noon Eastern)
15:28 JBoyer Snapshot
15:28 JBoyer Open Bugs - 2617
15:28 JBoyer Pullrequests - 71
15:28 JBoyer Signedoff - 42
15:28 JBoyer Updates Since Last Meeting
15:29 JBoyer Bugs Added - 40
15:29 JBoyer Pullrequest tag Added - 19
15:29 JBoyer Signedoff tag Added - 45
15:29 JBoyer Fix Committed - 52
15:29 JBoyer #topic New Business
15:29 JBoyer #info terranm89 would like some feedback on her LP tag proposal. :)
15:30 gmcharlt I've added a couple bits of feedback in the doc, but I'm overall +1 to the idea
15:30 terranm89 I've spoken with the CAT IG and with Acq  folks and they are all excited about easier tagging.
15:30 terranm89 There's not a way to batch-change LP tags is there? (I don't see any way with my permission level anyhoo.)
15:30 terranm89 gmcharlt++ thanks for the feedback
15:30 Dyrcona I have a question about the Lp tags changes: Why differentiate between needsrebase and needsrepatch?
15:31 JBoyer There would definitely be a difference in the amount of expected effort.
15:31 terranm89 In my head, needsrebase would be easier / require less actual code fixing
15:31 Dyrcona I don't necessarily think so, unless by needsrepatch we mean it doesn't work as advertised.
15:31 terranm89 That's how I was interpreting needsrepatch
15:32 JBoyer And I think there are a couple instances where that's the case.
15:32 terranm89 This is the description in the official tags list: testing revealed that the submitted code needs more work
15:32 terranm89 Of course, we could just change that description, too
15:33 Dyrcona I'm not sure that I like the two tags being so similar, but guess I'll have to live with it.
15:33 Dyrcona Also, I agree with gmcharlt regarding the deprecation tag.
15:34 terranm89 Whatabout "needsrebase" and "needswork" ?
15:35 terranm89 I've removed "deprecation" from the list of things to be deprecated :)
15:36 JBoyer You see, now we're talking about recursion...
15:36 terranm89 heh
15:37 Dyrcona needswork seems better. I don't want to get too hung up on terminology though. Tags are mainly useful for searching, and I pretty much only look for signedoff with Fix Committed unchecked. :)
15:37 jeff "outdated" and "broken"? :-)
15:37 JBoyer jeff++
15:37 terranm89 jeff++ lol
15:37 csharp_ "bustedyoufixit"
15:37 terranm89 heh
15:39 JBoyer Given how many characters match between needsrebase and needsrepatch I do kind of like changing repatch to needswork.
15:39 JBoyer Easier to tell the difference at a glance.
15:40 Dyrcona Actually, "broken" might be better. It's shorter.
15:41 Dyrcona I guess needswork is more polite.
15:42 JBoyer swingandamiss
15:42 terranm89 I kind of like having the needswork and needsrebase because they are alphabetically grouped with needstest and needsdiscussion - all things that are asking for some specific type of action
15:42 Dyrcona Yeah. I concur.
15:42 shulabear terranm89++
15:43 Dyrcona terranm89++
15:43 abneiman terranm89++
15:43 JBoyer terranm89++
15:43 abneiman indeed, great list terranm89
15:43 Bmagic #info Bmagic = Blake GH, MOBIUS, super late
15:43 terranm89 (and needsreleasenote)
15:43 gmcharlt terranm89++
15:44 terranm89 Okay, thanks everyone! I don't suppose there's a way to batch change any of these in LP?
15:44 terranm89 If not I'll recruit some folks from the interest groups to help
15:45 csharp_ terranm89's need put R.E.M.'s Pop Song '89 in my head :-)
15:45 Bmagic renaming them doesn't bring all the bugs with it?
15:45 csharp_ terranm89: no batching that I'm aware of
15:46 Dyrcona terranm89: There's some Python libraries that could be used, but I haven't messed with them for a while.
15:46 Dyrcona Seems to me there was a limit to how many things you could manipulate at once, too.
15:46 csharp_ Dyrcona: last time I looked at those they mostly didn't work as far as I could tell
15:46 abneiman terranm89: I'm happy to help, just not til next week
15:46 csharp_ probably not updated to python3 too
15:46 Dyrcona csharp_: There's a newer one on Ubuntu 20.04 that seems to work, even has a shell mode.
15:47 csharp_ Dyrcona: oh cool - I haven't looked in a while
15:47 JBoyer Bmagic, it doesn't look like you can rename them, just add new and hide the old.
15:47 terranm89 yes, was just typing that
15:47 terranm89 No worries - it's not a bad idea for people to review them and add additional tags at the same time anyway
15:48 JBoyer But it's possible those python libs might be able to help since the API may be able to do things we can't. Or at least it won't get bored doing it the same way we would. :)
15:48 Bmagic JBoyer: wow, that'll be "fun"
15:48 Dyrcona csharp_: lptools and python3-launchpadlib
15:49 terranm89 It's something I can do while I'm stuck in conference sessions all next week
15:50 JBoyer If there are no more LP comments I'll move along
15:50 csharp_ Dyrcona: thanks, I'll take a look
15:51 JBoyer csharp_, Dyrcona , if either of you two come up with something that does the job it might be good to throw something about it on the wiki.
15:51 JBoyer #topic Announcements
15:51 JBoyer #info Next Meeting is November 09, 2021
15:52 JBoyer And we're out.
15:52 JBoyer #endmeeting
15:52 pinesol Meeting ended Tue Oct 12 15:52:07 2021 US/Eastern.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)
15:52 pinesol Minutes:        http://evergreen-ils.org/meetings/evergr​een/2021/evergreen.2021-10-12-15.02.html
15:52 pinesol Minutes (text): http://evergreen-ils.org/meetings/evergr​een/2021/evergreen.2021-10-12-15.02.txt
15:52 pinesol Log:            http://evergreen-ils.org/meetings/evergree​n/2021/evergreen.2021-10-12-15.02.log.html
15:52 Bmagic JBoyer++
15:52 terranm89 JBoyer++
15:52 Dyrcona JBoyer++
15:52 shulabear jboyer++
15:52 abneiman JBoyer++
15:52 phasefx JBoyer++
15:52 csharp_ JBoyer++
15:55 JBoyer shulabear++
15:55 JBoyer terranm89++
15:55 JBoyer gmcharlt++
15:56 Dyrcona JBoyer: I had some scripts for updating Lp bugs by series for releases, setting Fix Committed to Fix Released, but they stopped working a few years ago. I have meant to revisit them.
16:03 JBoyer Dyrcona++ hopefully that newer python module will make that easier
16:04 jvwoolf joined #evergreen
18:00 pinesol News from qatests: Testing Success <http://testing.evergreen-ils.org/~live>

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat