| Time |
Nick |
Message |
| 00:06 |
pinesol_green |
[evergreen|Kathy Lussier] Docs: 2.11.9 release notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=203dd19> |
| 00:06 |
pinesol_green |
[evergreen|Kathy Lussier] Docs: 2.12.6 Release Notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=b38b3e5> |
| 00:24 |
|
acautley joined #evergreen |
| 01:46 |
|
acautley joined #evergreen |
| 02:55 |
|
acautley joined #evergreen |
| 03:49 |
|
acautley joined #evergreen |
| 04:37 |
|
Jillianne joined #evergreen |
| 04:58 |
|
acautley joined #evergreen |
| 05:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 05:55 |
|
acautley joined #evergreen |
| 06:59 |
|
acautley joined #evergreen |
| 07:13 |
|
rjackson_isl joined #evergreen |
| 08:00 |
|
acautley joined #evergreen |
| 08:05 |
|
_bott_ joined #evergreen |
| 08:24 |
|
collum joined #evergreen |
| 08:26 |
|
finnx joined #evergreen |
| 08:27 |
|
_bott_ joined #evergreen |
| 08:45 |
|
mmorgan joined #evergreen |
| 08:55 |
|
acautley joined #evergreen |
| 08:59 |
|
bos20k joined #evergreen |
| 09:10 |
|
kmlussier joined #evergreen |
| 09:12 |
bshum |
gmcharlt (and others): FYI, I'll angle to do another i18n POT and PO sync later tonight after today's work is done prior to 3.0-beta2 cutting (tomorrow?) |
| 09:14 |
|
Dyrcona joined #evergreen |
| 09:14 |
bshum |
Looking at LP translations, looks like Eva's finished translating for Czech all the latest strings we added for 3.0-beta1 for webstaff/db.seed so there's plenty to test and check out |
| 09:28 |
|
yboston joined #evergreen |
| 09:59 |
|
acautley joined #evergreen |
| 10:08 |
|
rlefaive_ joined #evergreen |
| 10:08 |
miker |
csharp: your recursion error is probably because of a loop in the environment object or target object of the event, post fleshing |
| 10:08 |
gmcharlt |
by overwhelming vote, it shall be named OpenSRF 3.0 |
| 10:09 |
gmcharlt |
bshum: noted, and thanks |
| 10:09 |
kmlussier |
Oops. The 2nd time this week I forgot to vote. |
| 10:18 |
csharp |
miker: pretty sure at this point that RAM was part of (if not the cause of) the problem - thanks for the pointer! |
| 10:18 |
Stompro |
Beta2 tomorrow... I'm just finishing with a test database upgrade of beta1... sigh. |
| 10:18 |
csharp |
Stompro: right there with ya |
| 10:19 |
csharp |
JBoyer's disable ALL THE TRIGGERS! solution made mine pretty painless though after sifting out PINES database particularities |
| 10:20 |
Stompro |
No problems with a 2.10.6 -> 3.0.Beta1 database upgrade so far though, so that is good. Some parts take a long time on my non ssd test vm though. |
| 10:21 |
Stompro |
csharp, could you point me to more info on that? |
| 10:26 |
csharp |
Stompro: here's what I did http://git.evergreen-ils.org/?p=evergreen/pines.git;a=blobdiff;f=Open-ILS/src/sql/Pg/version-upgrade/2.12.5-3.0-beta1-upgrade-db.sql;h=3a4f54fecc0b4c37b627da2963fe1e50e259aa02;hp=0da18594c992f7050b0ea907600e318af117d6da;hb=99d5c1164f7180117760d48245db54a5a7bc0296;hpb=dfc8e3892d76af32162e1069c2c620139449ce2f |
| 10:28 |
Stompro |
csharp++ thanks |
| 10:28 |
kmlussier |
IIRC, berick suggested that we incorporate the drop of triggers in the upgrade script. I think that would be a good idea. |
| 10:31 |
|
mdriscoll joined #evergreen |
| 10:32 |
csharp |
Stompro: also http://git.evergreen-ils.org/?p=evergreen/pines.git;a=commit;h=826d5ba27d47123982a9fc06b44847cca125c24e |
| 10:33 |
csharp |
excluding deleted recs may help too |
| 10:33 |
csharp |
kmlussier: agreed |
| 10:33 |
miker |
csharp: as in, bad stick of RAM? |
| 10:33 |
* miker |
hearts ECC |
| 10:34 |
csharp |
miker: as in "half the RAM I need" ;-) |
| 10:34 |
miker |
ha! ah, well then... |
| 10:34 |
csharp |
when we provisioned the new VM we did it on an older spec |
| 10:34 |
Stompro |
csharp, Thanks I'll try that out if the update hasn't completed in a few hours, after I'm done with a meeting. |
| 10:35 |
* csharp |
just noticed that WHERE NOT DELETED should be WHERE NOT deleted |
| 10:35 |
* Dyrcona |
loves his new DB server with 768GB of RAM and NVMe storage for the DB. |
| 10:35 |
csharp |
but it did work, so maybe that doesn't matter? |
| 10:36 |
* Dyrcona |
also happens to be working on the 2.12 db upgrade scripts, some parts of it take a long time, too. ;) |
| 10:36 |
kmlussier |
Recalculating fingerprints? |
| 10:38 |
Dyrcona |
Yeap. |
| 10:39 |
Dyrcona |
csharp++ # For the links.. |
| 10:40 |
Dyrcona |
I'm supposed to set up a 3.0 testing server, too, but I'm focusing on the upgrade, first. :) |
| 10:40 |
Dyrcona |
miker: Do you have any code for that set replication suggestion from last week? I might give that a go, though I said it made me uneasy. |
| 10:41 |
Dyrcona |
If not, I suppose I could figure it out. |
| 10:43 |
miker |
Dyrcona: it's basically just, "set session_replication_role to 'replica';" before the big update, and then "set session_replication_role to 'origin';" after |
| 10:44 |
Dyrcona |
miker: Thanks! I'll give that a whirl when do it the 3.0 upgrade on the testing server. |
| 10:44 |
pinesol_green |
[opensrf|Ben Shum] LP#1708048: Add support for Debian 9 Stretch - <http://git.evergreen-ils.org/?p=OpenSRF.git;a=commit;h=a85132e> |
| 10:44 |
miker |
awesome |
| 10:44 |
pinesol_green |
[opensrf|Jason Stephenson] LP#1708048: Fix ld problems by renaming libraries. - <http://git.evergreen-ils.org/?p=OpenSRF.git;a=commit;h=cace46d> |
| 10:44 |
pinesol_green |
[opensrf|Galen Charlton] bump up version numbers in master to reflect 3.0 as next major release branch - <http://git.evergreen-ils.org/?p=OpenSRF.git;a=commit;h=f3775e2> |
| 10:45 |
Dyrcona |
gmcharlt++ |
| 10:45 |
Dyrcona |
Should someone whip up a branch to remove Wheezy support? I feel like I should have done that earlier. :) |
| 10:48 |
pinesol_green |
[evergreen|Ben Shum] LP#1708048: Add support for Debian 9 Stretch - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=ca8ff2d> |
| 10:48 |
pinesol_green |
[evergreen|Jason Stephenson] LP#1708048: Fix ld issues by renaming libraries. - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=c8b8014> |
| 10:48 |
pinesol_green |
[evergreen|Galen Charlton] LP#1708048: mention Stretch support in upgrade doc - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=5eec50e> |
| 10:49 |
miker |
booooo |
| 10:49 |
miker |
wheezy forever! |
| 10:50 |
gmcharlt |
ok, miker will personally do Debian ETS for Wheezy! |
| 10:50 |
gmcharlt |
(Etneral Term Support) |
| 10:50 |
gmcharlt |
;) |
| 10:50 |
miker |
:) |
| 10:56 |
pinesol_green |
[evergreen|Galen Charlton] LP#1708048: bump up minimum required OpenSRF version - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=15678ae> |
| 10:58 |
Dyrcona |
:) |
| 10:59 |
Dyrcona |
Well, we typically only support two releases at once and Wheezy community LTS ends next Spring? |
| 11:02 |
|
rlefaive joined #evergreen |
| 11:09 |
|
Christineb joined #evergreen |
| 11:19 |
Dyrcona |
bshum: Lp 1718459 |
| 11:19 |
pinesol_green |
Launchpad bug 1718459 in OpenSRF "Remove Installation Targets for Debian Wheezy" [Undecided,New] https://launchpad.net/bugs/1718459 - Assigned to Jason Stephenson (jstephenson) |
| 11:30 |
|
rlefaive joined #evergreen |
| 11:35 |
gmcharlt |
tossing in bug 1715503 as ready for review and merging today |
| 11:35 |
pinesol_green |
Launchpad bug 1715503 in Evergreen "Complain if minimum Pg version for 3.0 is not met" [Medium,New] https://launchpad.net/bugs/1715503 |
| 11:40 |
|
rlefaive joined #evergreen |
| 12:46 |
bshum |
miker++ # I'll test your branch for bug 1717777 later tonight |
| 12:46 |
pinesol_green |
Launchpad bug 1717777 in Evergreen "Need to load all the PO files for a locale" [Undecided,Confirmed] https://launchpad.net/bugs/1717777 |
| 12:47 |
miker |
bshum++ # thanks |
| 12:47 |
bshum |
Fwiw, I also noticed that there could be differences in the various PO files too |
| 12:47 |
miker |
yeah ... but, I'm not to concerned about those, TBH |
| 12:47 |
bshum |
Once I merged all the Czech PO files together, I got a lot of fuzzy mismatches |
| 12:47 |
bshum |
Where they translated something like "Cancel" differently from one file to the next |
| 12:47 |
miker |
much more important that we not drop PO files on the floor by shoving the list into a hash ;) |
| 12:48 |
|
acautley joined #evergreen |
| 12:48 |
bshum |
My nextgen goal might to a plan to merge all the PO templates together and start eliminating those kind of discrepancies and removing redundant translation efforts by having the same strings translated multiple times per PO file |
| 12:48 |
bshum |
Maybe for 3.1, heh |
| 12:49 |
bshum |
Yeah, definitely looks like a good step in the right direction |
| 12:49 |
|
khuckins joined #evergreen |
| 12:57 |
|
remingtron joined #evergreen |
| 12:58 |
|
jihpringle joined #evergreen |
| 13:03 |
pinesol_green |
[evergreen|Galen Charlton] move significant upgrade considerations to one section in release notes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=be6edb3> |
| 13:17 |
|
collum joined #evergreen |
| 13:19 |
pinesol_green |
[evergreen|Galen Charlton] add acknowledgments dervied from Git history - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=2d91d94> |
| 13:27 |
|
bos20k joined #evergreen |
| 13:43 |
bshum |
Hmm, speaking of, I guess the QA server is running on Wheezy, right? |
| 13:44 |
bshum |
Something to think about as we work to remove Wheezy from makefile install targets, etc. |
| 13:45 |
Dyrcona |
Wheezy is old and...wheezy...it should be replaced by Jessie. :) |
| 14:10 |
Dyrcona |
FYI: Master Evergreen seems to be broken on Debian 8 Jessie. |
| 14:11 |
Dyrcona |
karma:unit starts out with this warning: WARN [watcher]: Pattern "/home/opensrf/Evergreen/Open-ILS/web/js/ui/default/staff/node_modules/angular-sanitize/angular-sanitize.min.js" does not match any file. |
| 14:12 |
Dyrcona |
Then all following tests blow up with with various long messages. |
| 14:13 |
Dyrcona |
I'm testing a branch that has this commit as head: bed0acd1b197af590da729b6e1c29839580d1c5b |
| 14:13 |
Dyrcona |
Oh, right. It doesn't look at the working repository. |
| 14:14 |
Dyrcona |
I may have botched a prereq in the makefile, but I don't think so. |
| 14:14 |
Dyrcona |
http://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/dyrcona/lp1718459-remove-debian7-wheezy |
| 14:14 |
|
jeff___ joined #evergreen |
| 14:15 |
Dyrcona |
I have to move on to something else right now, but will take another look later unless someone else does in the meantime. |
| 14:36 |
bshum |
asciidoc syntax problem in README when generating out the file for Evergreen master |
| 14:37 |
bshum |
They removed the word "Optional: " from the lead of the subheading for extra steps to install web client |
| 14:37 |
bshum |
But the bottom ---- underlining it wasn't shortenend too |
| 14:37 |
bshum |
This resulted in it misreading the heading and breaking all the subsequent sections too |
| 14:40 |
Dyrcona |
asciidoc-- |
| 14:43 |
pinesol_green |
[evergreen|Ben Shum] Docs: fixup asciidoc syntax for main install README - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=e4150bb> |
| 15:11 |
bshum |
I get Dyrcona's explosion too on a clean Jessie VM when trying to get through grunt all |
| 15:12 |
bshum |
I saw in the npm install a series of warnings about our node version being too old, so maybe the packaged nodejs in Jessie is no longer good enough for certain prereqs we are trying to pull in |
| 15:12 |
bshum |
Going to dig back at those and see what's the story there |
| 15:12 |
* bshum |
might suggest that we start using the nodejs LTS binary for all distros, not just the ones that don't come with packaged node |
| 15:13 |
bshum |
Since we're doing that with Stretch since Debian killed off Node packages |
| 15:22 |
bshum |
Dyrcona: Yep, if I yank out the prereqs for nodejs-legacy and npm for Jessie, and then install the NodeJS binary like we do for old-Wheezy, or Trusty, then run the usual npm install, grunt all is happy |
| 15:22 |
bshum |
So it is a problem with our package list and having old node on Jessie that's broken it for master |
| 15:23 |
gmcharlt |
and seems to have /just/ broken |
| 15:24 |
bshum |
I guess we should test it on Xenial where we also use the packaged nodejs and see if it's broken there too |
| 15:24 |
bshum |
gmcharlt: Extra fun! :) |
| 15:25 |
Dyrcona |
Yeah, we should test on Xenial and bug it, too. |
| 15:25 |
gmcharlt |
OK! fess up! who started using leftpad? ;) |
| 15:25 |
Dyrcona |
heh |
| 15:35 |
bshum |
Hmm, well the lovefield warning shows up in Xenial when doing the npm install |
| 15:35 |
bshum |
But it doesn't bomb the same way |
| 15:35 |
bshum |
So the nodejs legacy there isn't dying out |
| 15:35 |
bshum |
But I think we should just nix it and make very distro consistently use the same nodejs source |
| 15:40 |
Dyrcona |
I agree. |
| 15:40 |
Dyrcona |
The Debian Stretch release notes recommend using Node.js from source as they have no maintainers for the packages and there are often security issues with Node.js. |
| 15:42 |
jonadab |
I recommend not using Node.js, that solves THAT problem. |
| 15:42 |
jonadab |
Oh, wait... dependency... right. |
| 15:42 |
jonadab |
nvm. |
| 16:07 |
pinesol_green |
[evergreen|Bill Erickson] LP#1718301 Offline db connections via promise - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=b87a437> |
| 16:17 |
kmlussier |
berick++ |
| 16:29 |
* berick |
saw "leftpad", read it as "leafpad", started getting defensive |
| 17:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 17:03 |
|
genpaku joined #evergreen |
| 17:06 |
|
mmorgan left #evergreen |
| 17:17 |
* gmcharlt |
claims 1076 in the name of ghosts that are tired of being invisible everywhere! |
| 17:22 |
csharp |
well, RAM is not the solution to my A/T problem - back to log diving |
| 17:22 |
* csharp |
sighs in disgust |
| 17:25 |
csharp |
this is running OpenSRF 2.5.2/EG 2.12.4 |
| 17:25 |
* csharp |
seriously ponders rolling back to 2.11 just to have A/T back to kinda working |
| 17:26 |
csharp |
nothing in A/T changed right? why would this be suddenly so broken |
| 17:26 |
csharp |
acq EDI is dead in the water |
| 17:26 |
csharp |
and now notices are no longer reliably working |
| 17:28 |
csharp |
turning on debugging produces unusuably large log files |
| 17:29 |
berick |
csharp: acq sending or receiving? still using old-school edi (not edit attrs)? |
| 17:30 |
berick |
oh |
| 17:30 |
berick |
right |
| 17:30 |
berick |
becuase it's A/T |
| 17:30 |
berick |
duh |
| 17:30 |
jeffdavis |
any chance of a bug in 2.12.2 or later? we haven't had a/t issues on 2.12.1 |
| 17:30 |
berick |
jeffdavis: osrf 2.5.2 also? |
| 17:30 |
csharp |
berick: acq sending |
| 17:31 |
csharp |
for my notices issue (which is far more pressing) |
| 17:31 |
csharp |
2017-09-20 15:18:41 utility03 open-ils.trigger: [INFO:4899:Trigger.pm:864:] trigger: run_all_events firing group for grouped event def=112 and grp ident 2476480 |
| 17:31 |
csharp |
that's the last line before NOT CONNECTED TO THE NETWORK!!! started for that process |
| 17:31 |
csharp |
then it just starts consuming cstores and all is lost |
| 17:32 |
jeffdavis |
our OpenSRF is not-quite-2.5.1 |
| 17:33 |
jeffdavis |
rel_2_5 reset to a9cd124 |
| 17:33 |
pinesol_green |
jeffdavis: [opensrf|Bill Erickson] LP#1697029 Log and exit on write to dead child - <http://git.evergreen-ils.org/?p=OpenSRF.git;a=commit;h=a9cd124> |
| 17:33 |
miker |
csharp: any errors in the ejabberd logs? |
| 17:33 |
berick |
csharp: using A/T parallel processing? |
| 17:33 |
csharp |
sec... |
| 17:34 |
pinesol_green |
[evergreen|Mike Rylander] LP#1718300: Fix serial.unit visibility at creation time - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=8c04e4a> |
| 17:34 |
pinesol_green |
[evergreen|Galen Charlton] LP#1718300: stamp schema update - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=ab2edc4> |
| 17:34 |
pinesol_green |
[evergreen|Galen Charlton] LP#1718300: fold update into 2.12.5-3.0-beta schema update - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=3edbdfc> |
| 17:34 |
csharp |
berick: nope |
| 17:34 |
csharp |
we were before though - hmm |
| 17:34 |
csharp |
maybe I should try parallelizing |
| 17:35 |
* csharp |
runs off to find ejabberd logs |
| 17:35 |
miker |
csharp: /var/log/ejabberd/ |
| 17:36 |
csharp |
miker: nothing labeled "ERROR" and nothing that looks crazy to me |
| 17:41 |
csharp |
okay - turned on parallel processing 3 collect/3 react |
| 17:41 |
csharp |
will watch and see if it helps the opensrf issue |
| 17:42 |
csharp |
the acq A/T thing is almost certainly with the template/environment, but it's a real head scratcher because the template and env have not changed for literally years |
| 17:42 |
csharp |
same OS version same DB version |
| 17:43 |
csharp |
something in EG or OpenSRF must be different in a way that's harming it |
| 17:43 |
csharp |
unwinding that template ain't gonna be fun |
| 17:45 |
berick |
ok, so the acq issue is unrelated to the A/T issues |
| 17:46 |
pinesol_green |
[evergreen|Galen Charlton] LP#1715503: update install & upgrade doc to require Pg 9.4 - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=3eaf2e5> |
| 17:46 |
pinesol_green |
[evergreen|Galen Charlton] LP#1715503: make upgrade script abort if not running min support version - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=06cd847> |
| 17:46 |
pinesol_green |
[evergreen|Galen Charlton] LP#1715503: make eg_db_config refuse to create DB if min Pg version not met - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=5fc427c> |
| 17:48 |
pinesol_green |
[evergreen|Mike Rylander] Minor documentation adjustment for upgrade instructions - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=3e93cfc> |
| 17:48 |
csharp |
berick: right |
| 17:48 |
csharp |
just multiple hates on A/T for different reasons |
| 17:49 |
csharp |
right now that reason is opacity - it's very difficult to see what's going on, even with debugging on |
| 17:50 |
pinesol_green |
[evergreen|Cesar Velez] LP#1700635: enable eg-marc-editor for queued records in MARC Batch Import/Export - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=d07e882> |
| 17:54 |
jeffdavis |
csharp: is there any time lag between that run_all_events log message and the start of the "not connected to network" errors? |
| 17:56 |
csharp |
jeffdavis: looking... |
| 17:57 |
* berick |
was going to ask |
| 17:58 |
csharp |
jeffdavis: berick: nope - it's immediate |
| 17:58 |
csharp |
same second |
| 18:08 |
csharp |
2017-09-19 12:39:13 utility03 open-ils.trigger: [INFO:18023:Trigger.pm:864:] trigger: run_all_events firing group for grouped event def=114 and grp ident 3840911 |
| 18:08 |
csharp |
that was yesterday^^ again - same second |
| 18:21 |
|
acautley joined #evergreen |
| 18:39 |
|
Jillianne joined #evergreen |
| 18:40 |
bshum |
Bah |
| 18:40 |
bshum |
The xenial grunt test blows up on my latest VM rebuild test |
| 18:40 |
bshum |
So I guess that's all they wrote for that |
| 18:41 |
bshum |
I'll mock up my change branch for installing nodejs source and apply that in my test building |
| 18:44 |
bshum |
And in the meantime, while I wait for ansible to rebuild my test server (again), I'll go and make a bug for that change |
| 18:54 |
bshum |
Ho hum, might be a bigger problem than that.... sigh |
| 19:02 |
|
Dyrcona joined #evergreen |
| 19:02 |
bshum |
Well, clean repo copy, and still getting npm and dependency problems. Going to try nailing a copy of the errors this time around |
| 19:03 |
Dyrcona |
bshum: with the latest Node.js? |
| 19:04 |
bshum |
Dyrcona: Yep, I was getting failures, so I removed the legacy nodejs and npm from my xenial box and then tried using the latest NodeJS and it is still bombing on trying to grab all the stuff for web client |
| 19:04 |
bshum |
I'm not sure if it's NodeJS issues, or if it's changed that were merged last minute to master |
| 19:04 |
Dyrcona |
So the npm networks it likely hosed. |
| 19:04 |
bshum |
I was just thinking to go back a few commits to earlier in the evening when it was working last for me |
| 19:04 |
bshum |
Just to check |
| 19:04 |
Dyrcona |
OK. |
| 19:05 |
Dyrcona |
Do you have a branch with the changes, yet? |
| 19:05 |
Dyrcona |
It would make it easier for me to test, too. |
| 19:05 |
bshum |
For NodeJS? |
| 19:05 |
Dyrcona |
Yeah. |
| 19:05 |
bshum |
Yeah I was just adding the LP# and making the branch |
| 19:06 |
Dyrcona |
I could try with stretch, I guess. |
| 19:07 |
Dyrcona |
I've gotta resolve a conflict with master and my remove Wheezy branch. Someone updated the upgrade doc in the same way that I did. :) |
| 19:08 |
bshum |
Dyrcona: https://bugs.launchpad.net/evergreen/+bug/1718549 |
| 19:08 |
pinesol_green |
Launchpad bug 1718549 in Evergreen "install NodeJS from source for all supported distributions" [High,Triaged] |
| 19:08 |
Dyrcona |
bshum++ |
| 19:08 |
bshum |
I didn't touch the Wheezy stuff in my branch |
| 19:08 |
bshum |
But only changed the nodejs stuff, removed some old comments that aren't necessary, and then bumped the LTS NodeJS to the latest version |
| 19:09 |
bshum |
Let me go back to retrying my build though after backing out a few commits from earlier. I'm concerned about that last minute merge for offline DB |
| 19:09 |
bshum |
Just want to see if that's different, or if it really is a nodejs upstream problem |
| 19:11 |
Dyrcona |
OK. |
| 19:12 |
bshum |
Yep |
| 19:12 |
bshum |
Reverting b87a437825a531a3e6977d1e0b17e48d8f41f74a fixing my web client building |
| 19:12 |
pinesol_green |
bshum: [evergreen|Bill Erickson] LP#1718301 Offline db connections via promise - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=b87a437> |
| 19:12 |
bshum |
So something in that new commit broke nodejs fresh setup |
| 19:14 |
bshum |
Maybe all those nodejs changes that recently happened are unhappy with the change, but whatever gmcharlt/miker had installed during their testing was fine. |
| 19:15 |
bshum |
Either way, that's definitely the source of my breakage once I backed that back out |
| 19:15 |
* bshum |
resumes building out his test server for i18n stuff |
| 19:16 |
Dyrcona |
bshum: You typoed the Lp number in your branch, you swapped the 4 and 5. I noticed, 'cause you have my bug number from today. :) |
| 19:16 |
bshum |
Oops |
| 19:17 |
bshum |
That always happens to me :( |
| 19:17 |
gmcharlt |
bshum: hmm, I'l look at it after dinner |
| 19:17 |
gmcharlt |
maybe (hopefully) just a matter of tweaking a test |
| 19:18 |
bshum |
Dyrcona: I can fix and force push a replacement |
| 19:18 |
Dyrcona |
bshum: It's just the branch name no big deal for now. |
| 19:18 |
bshum |
Dyrcona: Ah okay, yes, I see that now. Doh. |
| 19:18 |
Dyrcona |
The commit messages have the correct number. |
| 19:19 |
bshum |
gmcharlt: Sounds good, I'll keep testing other stuff around it for now. Reverting it locally till I know what to do about it :) |
| 19:19 |
* bshum |
waits for the LONG i18n dance on his test server to play out |
| 19:21 |
|
jihpringle joined #evergreen |
| 19:23 |
gmcharlt |
https://www.youtube.com/watch?v=sArAC2_ow2k |
| 19:23 |
bshum |
gmcharlt: That helps actually, thanks :) |
| 19:24 |
* bshum |
looks for a Czech dancing video now |
| 19:35 |
bshum |
Alright, that multiple loading PO locales seems to be working out for me finally. So that's good. |
| 19:37 |
bshum |
Have to spend more time with the inner workings of those framed webstaff pages for acq, urlverify, reports, etc. to see what's going on with the locales there |
| 19:37 |
bshum |
But for fun, adding additional definitions like |
| 19:37 |
bshum |
PerlAddVar OILSWebLocale "cs_cz" |
| 19:37 |
bshum |
PerlAddVar OILSWebLocale "/openils/var/data/locale/url_verify/cs-CZ.po" |
| 19:38 |
bshum |
Adds the url_verify PO translations and that link checker starts showing with all the translated strings |
| 19:38 |
bshum |
If I add it to the location block for /eg/staff/ |
| 19:38 |
bshum |
In addition to the regular staff PO file for webstaff |
| 19:38 |
bshum |
But anywho |
| 19:38 |
bshum |
Patch looks good, I'll signoff and push it in for 3.0-beta2 happiness |
| 19:44 |
pinesol_green |
[evergreen|Mike Rylander] LP#1717777: absorb all PO files for a locale - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=41dc920> |
| 19:45 |
bshum |
I think I've spent too much time looking at the web staff client mostly in Czech. Using an English one looks foreign to me now... lol |
| 19:47 |
Dyrcona |
heh |
| 19:47 |
Dyrcona |
It sure lookes pretty in Armenian. :P |
| 20:06 |
|
jihpringle joined #evergreen |
| 20:28 |
|
acautley joined #evergreen |
| 20:49 |
gmcharlt |
Dyrcona: bshum: either of you around? |
| 20:49 |
bshum |
gmcharlt: I am |
| 20:50 |
gmcharlt |
k, got a patch for you to look at in a minute |
| 20:50 |
bshum |
Working my way through a Stretch VM install :) |
| 20:50 |
Dyrcona |
I'm here also. |
| 20:56 |
bshum |
Fwiw, using "grunt all --force" let me proceed through the whole install even with the failed stuff. So it does appear to be the tests like you said |
| 20:56 |
bshum |
I havent' tried installing Hatch and trying out everything else though |
| 20:56 |
gmcharlt |
user/gmcharlt/lp1718301_lf_no_multi_connect |
| 20:57 |
Dyrcona |
Well, given the message that I got earlier, I assumed a --force would fail to work in the end. |
| 21:01 |
bshum |
gmcharlt: That patch worked for me |
| 21:01 |
bshum |
I can sign and push it on |
| 21:01 |
Dyrcona |
Blows up for me on Jessie... I did a git clean -x -f -d. |
| 21:02 |
Dyrcona |
Probably should wipe out any /tmp directories. |
| 21:02 |
bshum |
Dyrcona: For Jessie, do you have the changes for using nodejs source? |
| 21:02 |
Dyrcona |
No. |
| 21:02 |
bshum |
That part still matters for Jessie I think. |
| 21:02 |
bshum |
I guess I'll go retest on Xenial just to be sure |
| 21:03 |
Dyrcona |
I've got the legacy packages installed. |
| 21:05 |
Dyrcona |
Well, I'm done for tonight, then. I'm too tired to remove the packages or wipe out the vm right now. |
| 21:08 |
Dyrcona |
With that, I'm signing out. |
| 21:09 |
gmcharlt |
bshum: OK, and offline is working as I expect - so yeah, please go ahead and signoff and merge that patch, it's at least part of the puzzle |
| 21:15 |
bshum |
Yep, Jessie is broken for me using the current nodejs version we install from apt sources |
| 21:16 |
gmcharlt |
so, broken exactly how? |
| 21:17 |
bshum |
Node complains a great deal about version being too low |
| 21:17 |
gmcharlt |
during the npm install? |
| 21:17 |
bshum |
And then it gets to a place where we get a series of angular stuff can't be found |
| 21:17 |
bshum |
Yeah, during npm install |
| 21:18 |
gmcharlt |
kk, trying something |
| 21:19 |
gmcharlt |
npm --version == 3.10.10? |
| 21:19 |
bshum |
1.4.21 |
| 21:19 |
* gmcharlt |
blinks |
| 21:19 |
bshum |
Like I said, something ancient, thanks a lot Jessie |
| 21:19 |
bshum |
It's installing nodejs-legacy and npm from Jessie's apt sources |
| 21:19 |
bshum |
Crazy old stuff |
| 21:20 |
bshum |
We *should* update it to install from source like we do for Debian Stretch and Ubuntu Trusty. |
| 21:21 |
gmcharlt |
DOH |
| 21:21 |
gmcharlt |
ok, I see why I am not having these problems on my jessie box |
| 21:21 |
gmcharlt |
'deb https://deb.nodesource.com/node_6.x jessie main' covers a multitude of sins |
| 21:21 |
gmcharlt |
so... |
| 21:21 |
bshum |
That's an option too I suppose |
| 21:22 |
gmcharlt |
do you have that install from source patch lined up for Jessie/Xenial? |
| 21:22 |
gmcharlt |
if so, I can grab it |
| 21:22 |
bshum |
gmcharlt: Yep, it's in https://bugs.launchpad.net/evergreen/+bug/1718549 |
| 21:22 |
pinesol_green |
Launchpad bug 1718549 in Evergreen "install NodeJS from source for all supported distributions" [High,Triaged] |
| 21:22 |
bshum |
The thing I'm testing right now is whether you have to do something first to remove old npm and nodejs-legacy too |
| 21:22 |
bshum |
And I think you might |
| 21:23 |
bshum |
Cause when I run my patch through, it installs new nodejs and I get node --version = 6.11.3 as I expect, but still getting npm --version of 1.4.21 |
| 21:23 |
bshum |
So not sure it's not acting funny |
| 21:23 |
bshum |
But for fresh installs, should go perfectly |
| 21:24 |
bshum |
Oh, could just be my environment |
| 21:24 |
gmcharlt |
which npm |
| 21:25 |
bshum |
3.10.10 now showing |
| 21:25 |
bshum |
So that means it overwrites correctly with the installed source |
| 21:25 |
bshum |
So that's fine then |
| 21:25 |
bshum |
And happily creating the web client now with my patch and yours for Jessie |
| 21:25 |
bshum |
I'll go ahead and push your change to master first gmcharlt |
| 21:25 |
bshum |
Thanks! |
| 21:26 |
gmcharlt |
kk |
| 21:26 |
gmcharlt |
(needless to say, in the cold light of tomorrow morning, we'll want to do some double-checking before dbwells starts turning the building crank) |
| 21:27 |
pinesol_green |
[evergreen|Galen Charlton] LP#1718301: catch it when multiple connection attempts fail - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=474afc4> |
| 21:27 |
bshum |
gmcharlt: Most definitely sounds like a good idea |
| 21:28 |
gmcharlt |
(on of the ironies is that users of the packages in principle don't need to care in the slightest) |
| 21:28 |
bshum |
True that |
| 21:31 |
bshum |
@later tell Dyrcona Ooooo speaking of Armenian, the "tpac" file for Armenian is 100% translated.... I know what I'm building a sample for next!! |
| 21:31 |
pinesol_green |
bshum: The operation succeeded. |
| 21:33 |
pinesol_green |
[evergreen|Ben Shum] LP#1718549: Use Nodejs source install for all distributions - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=cbeb079> |
| 21:33 |
pinesol_green |
[evergreen|Ben Shum] LP#1718549: Bump NodeJS version to v6.11.3 - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=85053a2> |
| 21:34 |
bshum |
gmcharlt: I'm going to spin up my i18n server to get a new POT and PO sync going for master; assuming things are a little more settled at the moment |
| 21:34 |
gmcharlt |
bshum: yeah, I think they're settled enough |
| 21:34 |
gmcharlt |
thanks! |
| 21:48 |
|
acautley joined #evergreen |
| 21:55 |
pinesol_green |
[evergreen|Ben Shum] Translation updates - newpot - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=d019d22> |
| 21:55 |
pinesol_green |
[evergreen|Ben Shum] Translation updates - po files - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=7a078f4> |
| 21:58 |
|
_bott_ joined #evergreen |
| 22:09 |
bshum |
Okay, xenial builds fine for me with the latest master as of all the POT/PO. I'll test further tomorrow. Thanks again gmcharlt for the patch |
| 22:09 |
bshum |
gmcharlt++ |
| 22:09 |
bshum |
Goodnight #evergreen :) |
| 22:17 |
gmcharlt |
https://evergreen-ils.org/evergreen-2-11-9-and-2-12-6-released/ |
| 22:17 |
gmcharlt |
Bmagic++ |
| 22:17 |
gmcharlt |
dbwells++ |
| 22:19 |
dbwells |
gmcharlt++ |
| 22:21 |
pinesol_green |
[evergreen|Dan Wells] Forward-port 2.11.9 upgrade script - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=d0636ba> |
| 22:25 |
pinesol_green |
[evergreen|Galen Charlton] forward-port 2.12.5-2.12.6 schema update - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=c202ac9> |
| 22:55 |
|
acautley joined #evergreen |
| 23:57 |
|
acautley joined #evergreen |
