Evergreen ILS Website

IRC log for #evergreen, 2014-07-08

| 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
00:53 dcook__ joined #evergreen
02:42 b_bonner_ joined #evergreen
02:43 mtcarlson_away joined #evergreen
05:06 pinesol_green Incoming from qatests: Test Success - http://testing.evergreen-ils.org/~live/test.html <http://testing.evergreen-ils.org/~live/test.html>
06:33 dbwells joined #evergreen
06:57 dbwells_ joined #evergreen
07:38 jboyer-isl joined #evergreen
08:24 mrpeters joined #evergreen
08:28 akilsdonk joined #evergreen
08:29 eeevil leftover egg roll ... that's the breakfast of champions, right?
08:31 mrpeters heck yeah, especially a chicken egg roll :)
08:31 mrpeters veggies, protien, what could be more healthy!
08:32 * csharp just read "chicken or the egg roll"
08:36 ericar joined #evergreen
08:37 adbowling-isl joined #evergreen
08:47 Shae joined #evergreen
08:48 collum joined #evergreen
08:56 jwoodard joined #evergreen
08:59 mmorgan joined #evergreen
09:00 tspindler joined #evergreen
09:04 jboyer-isl csharp: That's an easy one, egg rolls are an appetizer, chicken is the main dish. Ba-dum, pssh.
09:06 jeff serves me right for walking into a "chicken or the egg" joke before having my own breakfast.
09:10 csharp jboyer-isl: ++
09:10 csharp jboyer-isl++ # even
09:17 _bott_ joined #evergreen
09:24 yboston joined #evergreen
09:27 ericar_ joined #evergreen
09:37 RoganH joined #evergreen
09:54 Shae joined #evergreen
09:57 krvmga joined #evergreen
10:02 krvmga i got my 2.6.1 up and running successfully (yay!) and i installed the staff client and registered it successfully on two separate machines (again, yay!)
10:02 jboyer-isl krvmga++
10:03 krvmga but when i try to log in, the Loading data... starts up, retrieves printer settings and then gives me a network failure message.
10:03 krvmga method=open-ils.circ.circ_modifier.retrieve.all
10:03 krvmga params=[{"full":true}]
10:03 krvmga THROWN:
10:03 krvmga {"payload":[],"debug":"osrfMethodException : Method [open-ils.circ.circ_modifier.retrieve.all] not found for OpenILS::Application::Circ","status":404}
10:03 krvmga STATUS:
10:03 krvmga i'm not sure why it's going wrong at this point.
10:03 dcook joined #evergreen
10:04 Dyrcona joined #evergreen
10:04 jboyer-isl I've seen that before, but I forget what happened.
10:04 jboyer-isl Can you connect on the machine with srfsh?
10:04 krvmga yes
10:04 csharp heads up to mundungus admins - doing an OS update and reboot, FYI
10:04 jboyer-isl When was the last time you restarted apache? If it's cached old versions of one of the perl modules in memory that can cause issues.
10:05 krvmga i can try restarting apache
10:05 jboyer-isl Alternatively, you might check the logs for errors, apache should explain somewhere why it hadn't loaded the correct modules.
10:08 ningalls joined #evergreen
10:08 krvmga restarting apache didn't make a difference. checking log.
10:09 rjackson-isl joined #evergreen
10:10 ningalls joined #evergreen
10:11 jboyer-isl And if that doesn't work: Did you do all of the apache config steps? i.e. you have (ubuntu example) /etc/apache2/sites-enabled/eg.conf, /etc/apache2/eg_vhost.conf and /etc/apache2/eg_startup?
10:13 krvmga finished checking logs. nothing unusual. no errors
10:13 ericar_ joined #evergreen
10:14 krvmga jboyer-isl: i can surf to the site and browse the contents of the catalog
10:14 krvmga i've got those files in place
10:15 krvmga site enabled and so on
10:15 jeff it may simply be that your open-ils.circ service isn't running. i believe that will result in the 404 reported above.
10:15 jeff logs, or the --diagnostic option for osrf_control may shed some light (or at least confirmation)
10:17 krvmga jeff: i will check
10:21 ningalls joined #evergreen
10:21 krvmga well, that was informative.
10:21 krvmga lots of NOT configured for this host messages on osrf_control --diagnostic
10:22 tsbere krvmga: Perhaps you need the localhost argument?
10:23 krvmga tsbere: yes, all running for localhost
10:24 RoganH joined #evergreen
10:24 krvmga except opensrf.persist and opensrf.validator which have PIDs that match no running processs for either
10:24 tsbere krvmga: That would include on the diagnostic check (aka, osrf_control --localhost --diagnostic)
10:26 krvmga ERR opensrf.persist Has PID file entry [2973], which matches no running opensrf.persist processes
10:26 krvmga ERR opensrf.validator Has PID file entry [2981], which matches no running opensrf.validator processes
10:26 krvmga all the others report their uptimes
10:27 krvmga and their drone numbers
10:28 tsbere krvmga: Try running this and see what kind of output you get: perl -c -e 'use OpenILS::Application::Circ;'
10:30 krvmga as the opensrf user?
10:30 tsbere should probably work the same as pretty much any user
10:31 krvmga opensrf@libtrain:/openils/bin$ perl -c -e 'use OpenILS::Application::Circ;'
10:31 krvmga Can't locate Business/Stripe.pm in @INC (@INC contains: /etc/perl /usr/local/lib/perl/5.10.1 /usr/local/share/perl/5.10.1 /usr/lib/perl5 /usr/share/perl5 /usr/lib/perl/5.10 /usr/share/perl/5.10 /usr/local/lib/site_perl .) at /usr/local/share/perl/5.10.1/Op​enILS/Application/Circ/Money.pm line 30.
10:31 krvmga BEGIN failed--compilation aborted at /usr/local/share/perl/5.10.1/Op​enILS/Application/Circ/Money.pm line 30.
10:31 krvmga Compilation failed in require at /usr/local/share/perl/5.10.1​/OpenILS/Application/Circ.pm line 12.
10:31 krvmga BEGIN failed--compilation aborted at /usr/local/share/perl/5.10.1​/OpenILS/Application/Circ.pm line 12.
10:31 krvmga Compilation failed in require at -e line 1.
10:31 krvmga BEGIN failed--compilation aborted at -e line 1.
10:31 csharp krvmga: (as root) cpan Business::Stripe
10:32 * tsbere suspected that would be the issue, but the perl command verified it
10:33 krvmga tsbere: did that. successful. then
10:33 krvmga opensrf@libtrain:/etc/apache2$ perl -c -e 'use OpenILS::Application::Circ;'
10:33 krvmga -e syntax OK
10:34 tsbere krvmga: Restart the circ service (or everything) and you may be good to go
10:35 krvmga tsbere: is that osrf_control --stop and --start?
10:37 RoganH joined #evergreen
10:40 Bmagic joined #evergreen
10:41 krvmga brb. rebooting machine.
10:50 RoganH joined #evergreen
10:55 krvmga joined #evergreen
10:57 krvmga everything now working as expected :)
10:57 krvmga tsbere++
10:57 krvmga jeff++
10:58 krvmga jboyer-isl++
10:58 kbeswick joined #evergreen
10:59 vlewis joined #evergreen
11:04 ningalls joined #evergreen
11:35 rfrasur joined #evergreen
11:39 RoganH joined #evergreen
11:55 bmills joined #evergreen
12:02 b_bonner_ joined #evergreen
12:04 mtate joined #evergreen
12:04 RoganH joined #evergreen
12:08 ningalls joined #evergreen
12:15 b_bonner_ joined #evergreen
12:30 sseng on a brick setup, would anyone have any insight into why a pcrud.retrieve works but a pcrud.update would fail? pcrud.retrieve('bre', recId) would retrieve but pcrud.update(bib) won't update (marcedit.js)
12:34 dbs permissions? Or garbage MARCXML is getting submitted.
12:35 sseng dbs: interesting part, it works on a non brick setup
12:35 berick sseng: memcache settings in eg_vhost.conf
12:35 berick OSRFTranslatorCacheServer
12:36 berick all apache instances need to talk to the same cache server
12:37 sseng berickl: dbs: oh ok i do remember having to change memchache settings in the opensrf.xml, going to look
12:37 berick sseng: /etc/apache2/eg_vhost.conf
12:37 berick not opensrf.xml
12:38 sseng berick: got it. do these settings say, get erased when an accidental power outage happens?
12:40 bshum Seems unlikely, if it's written in the config files, I would expect those settings to be used when services are started again next.
12:41 sseng berick: got it
12:41 sseng bshum: got it
12:48 sseng bshum++ berick++ dbs++ : thanks a bunch!!! updating the OSRFTranslatorCacheServer to point to the same memcache server did the trick!
12:53 * bshum has been there before, berick++
12:57 ldw joined #evergreen
13:13 chatley joined #evergreen
13:16 tspindler1 joined #evergreen
13:19 dcook__ joined #evergreen
13:20 tspindler1 left #evergreen
13:22 RoganH joined #evergreen
13:31 RoganH joined #evergreen
14:10 mtcarlson joined #evergreen
14:10 b_bonner joined #evergreen
14:20 jeffdavis Browsing the catalogue is quite slow on our 2.6.1 system: 72208.924 ms for SELECT * FROM metabib.browse( 'title', 'dhalgren', '1', NULL, 'f', NULL, '10' );
14:20 jeffdavis Anyone else seeing response times like that?
14:29 mmorgan I get 626.089ms on a training server running 2.5.3 and 4489.063ms on production running 2.5.5.
14:29 mmorgan though, production's indexes aren
14:29 mmorgan 't complete yet.
14:30 yboston bshum: are you available for a moment?
14:32 bshum yboston: Sure thing, sorry I'm here now.
14:32 yboston bshum: I wrote out the release notes for that small feature
14:33 yboston bshum: should I commit it to the cataloging folder in the "next" section?
14:33 bshum Hmm, that sounds like a good place as any to me.
14:33 yboston bshum: not sure if the commit that it covers has been added yet, not sure if that really maters.
14:33 yboston (added to master)
14:33 bshum It hasn't been pushed to master yet.
14:34 bshum What I might suggest is taking your signoff branch and adding the release note to that branch
14:34 bshum And then pushing it to working
14:34 bshum From there, I can pick both the feature changeset and your notes into master
14:34 bshum With my signoff
14:34 yboston that makes sense
14:35 yboston I plan to do a fectch & merge (or pull) on that signoff branch. I wonder if I would need to do a rebase, never done one before
14:36 bshum It's up to you, always fun to experiment ;)
14:37 yboston would it be advisable NOT to do a fetch & merge (or pull) on that signoff branch? I assumed I should (trying to learn good Git practices and the reasons behind it)
14:42 bshum yboston: If it was me, I would only pick out the change I wanted in a new branch that I wanted to keep building on top of.
14:42 Dyrcona yboston: Best practice depends on whom you ask and what community you're talking about.
14:42 yboston Dyrcona: thanks, I should have been more clear, I meant the EG community
14:42 Dyrcona yboston: In Evergreen, we tend to use cherry-pick over merge. Other communities prefer merge.
14:43 Dyrcona yboston: Yeah, I figured.
14:43 yboston Dyrcona: thank you for bringing it up to make things more clear
14:44 Dyrcona yboston: This documentation for a new feature?
14:45 yboston bshum: I am not 100% sure what you mean. Part of the problem is that I suspect you gave me a suggestion that was the path of least resistance for my current situation.
14:46 yboston bshum: Do yo mean I should do a fresh branch off master and cherry pick Bil''s commit and and one of mine as an alternative of just adding a new commit to my sign off>
14:46 yboston ?
14:46 bshum yboston: So there's lots of ways to move ahead
14:46 bshum You could make a git branch based on your working branch that you pushed
14:46 Dyrcona yboston: if you do a merge, try it with the --ff-only option.
14:46 bshum Add new changes, commit it, and push again to the same working branch name, since it just builds on your existing chain
14:46 Dyrcona That will only merge fast forward changes or it will say it can't fast forward.
14:46 bshum If you cherry pick Bill's change with your signoff to a new master branch, you could do that too
14:46 bshum And then build on that
14:47 yboston Dyrcona: Bill had made a branch with a single commit that added a simple new feature. I just created a local commit just to add the release note
14:47 bshum (which isn't a bad way to do it either, since that'll spare me from having to rebase your commits onto master again)
14:47 yboston Dyrcona: I had made a signoff branch on working for bill's commit.
14:47 bshum That also would work, making your change in another branch entirely, and we pick from bill's branch and yours to get all the changes into one branch later.
14:47 bshum (so many paths to completion!)
14:47 bshum @love git
14:47 pinesol_green bshum: But bshum already loves git!
14:47 Dyrcona ok, then. is this a collab branch or a usr branch we're talking about? I mean bill's branch.
14:48 bshum It's a user branch.
14:48 Dyrcona Best thing would be to push to a new collab branch and indicate it in the launchpad bug.
14:49 yboston thanks for all the feedback, I am glad it all make sense at this point. I few moths ago not so much
14:49 yboston *months
14:51 Dyrcona We've all had to learn this stuff at some point, so we've all been there with the questions.
14:51 yboston for instance, with 20/20 hindsight I bet bill and I would have agreed to use a collab branch form the beguining, but my creation of the release ntoes was not planned. though it has been a great learning expereince
14:52 Dyrcona yboston: If you signed off on bill's commit, you'd still need another collab branch to push to.
14:52 Dyrcona When you sign off, the commit hash changes, and you would have to force push to the collab branch.
14:52 Dyrcona The way our server is set up, you can't force push to a branch that someone else owns.
14:58 yboston Dyrcona: thanks for the infoon the collab brnaches, I am reviewing some git stuff on my end before I ask more quesiotns, one sec
14:59 hbrennan joined #evergreen
15:15 yboston bshum: I think I will try to set up a new collab branch from master and 1) cherry pick bill's commit 2) my signoff commit 3) my release notes commit
15:15 yboston bshum: One reason to try this approach is to save each other form having to rebase, would that be OK?
15:15 bshum Sounds good.  For fun, 1 & 2 can be done at the same time with something like "git cherry-pick -s ######" (the -s signs off during the cherry-pick)
15:16 yboston that makes sense to combine those too steps with "-s"
15:16 Dyrcona git checkout -b new_branch origin/master
15:16 bshum Dyrcona++
15:17 yboston I will proceed, and leave my rebase expermiments for a little later
15:17 Dyrcona then do your cherry-pick(s) with -s...
15:18 yboston Dyrcona: thanks, of course I have to fetch & merge (or pull) before hand
15:18 Dyrcona yboston: only fetch. using origin/master its based on the remote branch in origin.
15:18 yboston Dyrcona: I see what I missed now
15:19 Dyrcona it's a good idea to keep your local master up to date though, so git pull in master is good if you want.
15:19 yboston I have been keepng my local master up to date
15:20 yboston but your tip is a good reminder how to quickly grab a fresh branch from the remote oririn/master no matter the curretn state of my local master
15:21 Dyrcona Another thing about git checkout -b new_branch :
15:22 Dyrcona If you don't specify a parent branch, it will use your currently checked out branch.
15:22 Dyrcona so, this would achieve the same result:
15:22 Dyrcona git checkout master; git fetch --all ; git pull ; git checkout -b new_branch
15:23 Dyrcona so, your new branch is based on your locally updated master at that point.
15:23 Dyrcona If that's too much information, tell me, and I'll stop. :)
15:28 ericar joined #evergreen
15:28 yboston Dyrcona: no this was helpful, it confrims my assumptiions on dufferent workflows
15:29 Dyrcona Glad to help.
15:29 Dyrcona I usually use whatever work flow my fingers remember first. :)
16:11 jwoodard joined #evergreen
16:19 yboston bshum: I just pushed a new collab brnach with Bill's change and my release notes. A big thank you to you and Dyrcona for all the great tips
16:30 Dyrcona yboston++
16:36 bshum yboston++ # I'll check it out in a bit.
16:38 Bmagic can anyone here tell me what they have in their config for pound proxy? I am encountering a cookie issue
16:42 mrpeters left #evergreen
17:05 bshum I'd help you Bmagic, but we don't employ poundproxy, still on ldirectord.  Hope you find something to solve your issues though.
17:08 jeff Bmagic: can you elaborate on the issue you're having?
17:08 Bmagic bshum: When opening a specific report, the staff client complains about cookies not being enabled
17:09 Bmagic There is a clause for cookies within the poundproxy config relating to cookies and it wants something like this: ID      "userid"
17:10 jeff we have no lines regarding cookies in our pound config.
17:18 Bmagic jeff: interesting, and I am assuming you don't get this error
17:18 Bmagic jeff: does pp always send the same client to the same app server once a session is established?
17:18 Bmagic jeff: Are your "backend"'s setup over port 80?
17:20 mmorgan left #evergreen
23:16 Bmagic joined #evergreen

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