| Time |
Nick |
Message |
| 00:06 |
bshum |
Well that's just weird.... modperl was enabled by default in my fresh install |
| 00:06 |
bshum |
I know we moved it out of opensrf and into Evergreen... maybe something is wonky for Trusty |
| 00:07 |
bshum |
Oh okay |
| 00:08 |
bshum |
We haven't finished the juggling for all that yet... |
| 00:09 |
* bshum |
strongly wants to see https://bugs.launchpad.net/opensrf/+bug/1579219 merged for OpenSRF 2.5.0 and also for Evergreen 2.12 then. |
| 00:09 |
pinesol_green |
Launchpad bug 1579219 in OpenSRF "don't require mod_perl as an OpenSRF dependency" [Wishlist,Triaged] |
| 00:09 |
* bshum |
updates targets accordingly |
| 00:09 |
bshum |
Well, for Evergreen. Looks like we don't have an OpenSRF 2.5.0 final yet. |
| 00:09 |
bshum |
For bug milestone |
| 00:12 |
bshum |
Guess I haven't built on Ubuntu 14.04 in awhile. I totally forgot about mod_perl oddness till now. |
| 00:13 |
bshum |
And huzzah, working Ubuntu 14.04 with PG 9.4 and OpenSRF master and Evergreen master (with the working branch changes) |
| 00:13 |
bshum |
Whee |
| 00:13 |
bshum |
Tests pass, and webclient functions. Good enough. |
| 00:15 |
* bshum |
starts Fedora 25 download and goes to bed |
| 04:28 |
|
NawJo joined #evergreen |
| 05:00 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 07:21 |
|
rjackson_isl joined #evergreen |
| 07:25 |
|
agoben joined #evergreen |
| 08:00 |
|
remingtron_ joined #evergreen |
| 08:16 |
|
collum joined #evergreen |
| 08:19 |
|
mdriscoll joined #evergreen |
| 08:30 |
|
dbwells_ joined #evergreen |
| 08:43 |
|
mmorgan joined #evergreen |
| 08:56 |
|
krvmga joined #evergreen |
| 09:18 |
csharp |
reports+- |
| 09:26 |
|
maryj joined #evergreen |
| 09:36 |
bshum |
Fedora 25's installer is neat. But I think package names must have drifted or something cause OpenSRF doesn't install prereqs. And dnf vs. yum? Hmm |
| 09:56 |
|
Dyrcona joined #evergreen |
| 09:57 |
|
terran joined #evergreen |
| 10:06 |
csharp |
berick: I've uncovered more data on the hold targeter issue - in all 6 examples I have, the "right" (i.e., expected) copy was targeted when the hold was placed, but was not the copy selected later (I'm assuming it's being changed by the nightly targeter run) |
| 10:06 |
dbs |
bshum: yeah, Fedora is in significant need of updates |
| 10:07 |
berick |
csharp: thanks. that should be consistent w/ the old hold targeter, right? |
| 10:07 |
csharp |
berick: I'm log-diving for what the nightly targeter run does (45 22 * * * . /etc/profile && /openils/bin/hold_targeter_v2.pl --parallel 3 --target-all) |
| 10:08 |
berick |
csharp: i can say for sure it will pick a different copy every time if it can. |
| 10:09 |
csharp |
berick: is there a way to say "leave ones with the right target alone"? |
| 10:09 |
* tsbere |
wonders how you define "right" in that case |
| 10:09 |
csharp |
(e.g., do I need --skip-viable?) |
| 10:09 |
berick |
well first and foremost, is it your thinking that this is new behavior? |
| 10:10 |
berick |
the intent is for a seemless migration to the new targeter |
| 10:10 |
csharp |
berick: we think so, I uncovered a third ticket about it that one of my colleagues closed |
| 10:11 |
berick |
ok, then i'm going to review the old targeter code, if confirmed, i'll push a patch. |
| 10:11 |
csharp |
we didn't receive any similar complaints before the upgrade (which is all very circumstantial) |
| 10:11 |
csharp |
berick: thanks - I'll keep looking too |
| 10:13 |
tsbere |
csharp/berick: From my POV, that sounds like what the existing hold targeter would have done. Just FYI. |
| 10:14 |
berick |
tsbere: k, thanks |
| 10:15 |
csharp |
berick: we're wondering if the previous check time is maybe getting overlooked? |
| 10:15 |
csharp |
it's supposed to leave the hold alone for 24h, right? |
| 10:15 |
tsbere |
*that* could be a difference, hadn't thought about the timing because no times were mentioned |
| 10:15 |
* csharp |
hasn't looked at the code yet - still collecting data & theories |
| 10:15 |
tsbere |
Though I wonder if that is what --target-all is doing? |
| 10:16 |
Dyrcona |
Yeah, prev_check_time has often been the answer to "weird" hold questions. |
| 10:16 |
berick |
oh wait! |
| 10:16 |
berick |
yes |
| 10:16 |
berick |
csharp: remove the --target-all flag |
| 10:16 |
csharp |
berick: ah - ok |
| 10:16 |
berick |
tsbere++ |
| 10:16 |
csharp |
tsbere++ berick++ |
| 10:17 |
berick |
--target-all says retarget all holds regardless of their last check time |
| 10:17 |
* tsbere |
hopes that it also skips those with a capture time ;) |
| 10:18 |
csharp |
whew! |
| 10:18 |
csharp |
thanks |
| 10:18 |
bshum |
dbs: Honestly I'm not sure we should keep Fedora notes in the official installer guide if we're this far out of date, and also not recommending Fedora for any actual production installations. :\ |
| 10:18 |
berick |
tsbere: yes, it still skips captured (and canceled, fulfilled, frozen) holds |
| 10:19 |
Dyrcona |
Yeah, I've sometimes thought that we should remove Fedora for that reason and because it isn't intended for production use. |
| 10:21 |
dbs |
bshum: makes sense to me |
| 10:21 |
berick |
csharp: your targeter should get a lot faster w/o --target-all as well |
| 10:23 |
csharp |
berick: awesomme |
| 10:23 |
csharp |
extra M awesome, so even awesommer |
| 10:24 |
* csharp |
's keyboard has begun repeating letters (M's and G's are the ones I've noticed) lately |
| 10:25 |
|
kmlussier joined #evergreen |
| 10:27 |
dbs |
bshum: as they are doc changes, I can just push the corresponding deletions for opensrf and evergreen |
| 10:32 |
dbs |
b397106c15c for OpenSRF master |
| 10:32 |
|
mmorgan1 joined #evergreen |
| 10:34 |
* dbs |
sees mention of PostgreSQL 9.1 in Evergreen README in master, ugh |
| 10:34 |
bshum |
dbs: Yeah, we're working on changing all that |
| 10:35 |
dbs |
bshum: right, I saw the discussion yesterday, but seeing the minimum of 9.3 and then "but Trusty comes with 9.1 so you're all good" a few paragraphs later o.0 |
| 10:35 |
pinesol_green |
[opensrf|Dan Scott] Remove Fedora as a well-tested target - <http://git.evergreen-ils.org/?p=OpenSRF.git;a=commit;h=b397106> |
| 10:36 |
dbs |
okay, I'll limit my edits to just getting rid of Fedora mentions and leave the postgresql-related updates to someone else |
| 10:39 |
bshum |
dbs: Fwiw, I started updates for that language in the collab for PG 9.4 support |
| 10:39 |
Dyrcona |
Trusty comes with 9.3. |
| 10:39 |
bshum |
http://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/collab/bshum/PG94-testing has the latest |
| 10:40 |
Dyrcona |
I think that was a typo. :) |
| 10:40 |
Dyrcona |
I'm testing that branch with Trusty, now. |
| 10:41 |
Dyrcona |
We have it so 9.4 gets installed automagically from apt.postgresql.org. |
| 10:45 |
pinesol_green |
[evergreen|Dan Scott] Remove Fedora from install documentation - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=9389188> |
| 10:46 |
miker |
berick / csharp: and to stick a fork in it, yes, the old targetter would move the hold to a new target upon (prev_check_time-exceeded) retargetting. line ~1390 of Storage/Publisher/action.pm ... |
| 10:53 |
gmcharlt |
http://libmail.georgialibraries.org/pipermail/open-ils-dev/2017-March/010309.html |
| 10:55 |
berick |
miker++ thanks for confirming |
| 10:55 |
bshum |
dbs++ # thanks |
| 10:56 |
csharp |
miker++ |
| 11:05 |
Bmagic |
I've never seen this error before. egweb: Context Loader error: Bad arg length for Socket::pack_sockaddr_in length is 0, should be 4 at /usr/lib/x86_64-linux-gnu/perl/5.22/Socket.pm line 156. |
| 11:06 |
Bmagic |
this is a fresh installation, I must have missed a configuration step somewhere |
| 11:11 |
|
sandbergja joined #evergreen |
| 11:11 |
csharp |
miker: thanks for pointing out bug 1174498 - I didn't find it when searching for dupes on bug 1669443 |
| 11:11 |
pinesol_green |
Launchpad bug 1174498 in Evergreen "Payment by billing type breakdown" [Wishlist,Triaged] https://launchpad.net/bugs/1174498 |
| 11:11 |
pinesol_green |
Launchpad bug 1174498 in Evergreen "duplicate for #1669443 Payment by billing type breakdown" [Wishlist,Triaged] https://launchpad.net/bugs/1174498 |
| 11:14 |
miker |
csharp: np ... it's been simmering in the back of my mind for years :) |
| 11:18 |
|
khuckins_ joined #evergreen |
| 11:23 |
berick |
Bmagic: just curious, what OS? |
| 11:23 |
Bmagic |
berick: xenial but I think I figured it out, Misconfig for memcached |
| 11:23 |
csharp |
Bmagic: are you running your branch for bug 1174498 in production? |
| 11:23 |
pinesol_green |
Launchpad bug 1174498 in Evergreen "Payment by billing type breakdown" [Wishlist,Triaged] https://launchpad.net/bugs/1174498 |
| 11:23 |
berick |
Bmagic: *nod* glad you figured it out |
| 11:24 |
|
brahmina joined #evergreen |
| 11:29 |
Bmagic |
berick: but yeah, we are running that branch in production. Why do you ask? |
| 11:30 |
berick |
csharp: ^-- |
| 11:30 |
csharp |
Bmagic: I asked :-) - because I'd love to see some payment/billing linkages |
| 11:30 |
csharp |
I'll add that to my testing for 2.12 once we get it running on our staging servers |
| 11:31 |
Bmagic |
so far so good. Our libraries seem to enjoy it |
| 11:31 |
csharp |
I bet |
| 11:32 |
Bmagic |
count(csharp) == count(berick) = glance cross confuse, lol |
| 11:32 |
csharp |
Bmagic++ |
| 11:33 |
|
kmlussier joined #evergreen |
| 11:34 |
|
mmorgan joined #evergreen |
| 11:57 |
Dyrcona |
Bmagic: Most languages will give you an illegal assignment error... ;) |
| 11:58 |
Bmagic |
Dyrcona: are you responding to my issue above "h for Socket::pack_sockaddr_i...." ? |
| 12:02 |
berick |
kmlussier: FYI @ bug 1669519 |
| 12:02 |
pinesol_green |
Launchpad bug 1669519 in Evergreen "Support seamless duplicate (or more) holds for anointed recipient groups" [Wishlist,New] https://launchpad.net/bugs/1669519 |
| 12:07 |
|
jihpringle joined #evergreen |
| 12:30 |
Stompro |
miker++ benchmarks++ |
| 12:51 |
|
yboston joined #evergreen |
| 13:01 |
Dyrcona |
kmlussier: Thoughts on lp 1493824? I wanted to hear from you, as RM, before pushing the branch. |
| 13:01 |
pinesol_green |
Launchpad bug 1493824 in Evergreen "Evergreen/PostgreSQL 9.4 support" [Wishlist,Confirmed] https://launchpad.net/bugs/1493824 |
| 13:02 |
* kmlussier |
looks more closely |
| 13:04 |
kmlussier |
Dyrcona: Sure, go for it. Since we most likely will be recommending 9.4 for 2.12, it's probably something that should be there. |
| 13:04 |
Dyrcona |
OK. Thanks. I'll push it. |
| 13:06 |
pinesol_green |
[evergreen|Ben Shum] LP#1493824: Install PG 9.4 on Debian Wheezy and Ubuntu Trusty - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=4f6424c> |
| 13:13 |
kmlussier |
Dyrcona++ bshum++ csharp++ |
| 13:15 |
* Dyrcona |
thinks he should have spent an extra five minutes proofreading the commit message, but oh well. |
| 13:15 |
* kmlussier |
read the commit message and didn't see anything wrong with it. |
| 13:19 |
Dyrcona |
Well, I have lidbdi in one place and install should be installing in another, but anyway. |
| 13:51 |
csharp |
Dyrcona++ bshum++ # pushing forward |
| 13:52 |
csharp |
the only concern (voiced by gmcharlt, I believe) was whether it was OK to more or less silently add the repo and repo keys |
| 13:54 |
Dyrcona |
Yeah. |
| 13:54 |
Dyrcona |
But, hey! With systemd, everything is more like Windows, now! :) |
| 13:55 |
bshum |
I guess we could have just linked to the apt source steps and let the reader go to the PG community's page on how to do it and leave it up to them to figure out. But ease of use isn't always so bad in my book |
| 13:56 |
* dbs |
notes that the ".service" is no longer necessary on Fedora or Debian for systemctl commands, hopefully Ubuntu follows |
| 13:57 |
dbs |
So, MARC Batch Import: when importing all from a queue, is it normal for the queue to just die so you have to go back to Inspect Queue and then select a set of the non-imported records and import again? |
| 13:57 |
dbs |
(2.10 here) |
| 13:58 |
kmlussier |
I wouldn't say it's normal, but MARC Batch Import does tend to have issues with large volumes of records. How large is the queue? |
| 13:58 |
dbs |
kmlussier: this was just 227 records, it imported the first 122 and then died |
| 13:59 |
kmlussier |
No, that doesn't sound normal to me. But I could also ask the person who actually does importing on a regular basis. |
| 14:00 |
rhamby |
back when I was doing acq stuff and importing record batches routinely I didn't have issues with batches of that approximate size (2.7/2.9 at the time) |
| 14:10 |
dbs |
yeah, weird, no failures reported in the queue. oh well, I normally import via SQL anyway so I'm not going to sweat it |
| 14:13 |
Dyrcona |
dbs: Mabye record 123 is bad? |
| 14:13 |
Dyrcona |
Maybe, even. |
| 14:14 |
dbs |
Dyrcona: if it was bad, I would expect a failure to have been reported. Or for it to not import when I try it again |
| 14:17 |
Dyrcona |
dbs: True. |
| 14:18 |
Dyrcona |
It has been a while since I used import directly, but I never had it just fail like that. |
| 14:18 |
Dyrcona |
The client would time out, but the import would continue anyway. |
| 14:19 |
Dyrcona |
I usually use a Perl script or SQL on large batches. |
| 14:19 |
* Dyrcona |
is not a cataloguer. |
| 14:21 |
* dbs |
was trying to go through the steps to enable technicians to use the friendly interface but... maybe in 2.12 |
| 14:23 |
Dyrcona |
How many records is too many can vary based on load on the servers, etc. |
| 14:27 |
Dyrcona |
That gives me 181 "Library" and 1 "Deparment." |
| 14:27 |
Dyrcona |
meh. you can insert the t. ;) |
| 14:27 |
Dyrcona |
Heh. wrong window... |
| 14:27 |
Dyrcona |
Not only can't I type, I can't mouse, either. |
| 14:27 |
* Dyrcona |
needs a nap. :) |
| 14:28 |
dbs |
Dyrcona++ |
| 14:29 |
dbs |
this batch of 89,983 records is not going to be added via the staff client |
| 14:37 |
|
rlefaive joined #evergreen |
| 14:39 |
Dyrcona |
I've been doing batches in the 100s of thousands lately. |
| 14:39 |
Dyrcona |
Usually pre-split into batches of 10,000 or so each. |
| 14:40 |
Dyrcona |
That's how they send 'em to me. |
| 14:52 |
kmlussier |
dbs: I checked with the person who imports records on a regular basis, and she concurs that it's unusual for it to die with a queue of that size. |
| 15:02 |
|
remingtron joined #evergreen |
| 15:32 |
|
Jillianne joined #evergreen |
| 15:37 |
|
jihpringle joined #evergreen |
| 15:56 |
Dyrcona |
Does anyone have a patch for edi mapper.rb to put the PO name in the pseudo-magical spot? |
| 15:56 |
Dyrcona |
I've installed berick's GIR-segments-for-copy-data branch, but apparently that isn't the thing. |
| 15:57 |
Dyrcona |
Someone keeps tell me to compare old files with new, but the old files are a server that won't boot, about 50 miles from where I now sit. |
| 15:58 |
Dyrcona |
Let this be a lesson: Don't check work email when you're on vacation. |
| 15:59 |
Dyrcona |
IDK. Mabye I can figure it out, but ruby.... |
| 16:01 |
Dyrcona |
Yeah, no. It's all ruby to me. |
| 16:08 |
Dyrcona |
So, it's not the ruby. I don't know EDI. |
| 16:10 |
jeffdavis |
Dyrcona: possibly in https://github.com/BCLibCoop/openils-mapper ? |
| 16:11 |
jeffdavis |
although I don't think we have much more there than what's in berick's version |
| 16:13 |
Dyrcona |
jeffdavis++ |
| 16:14 |
Dyrcona |
I think it's there, actually. I'm going to clone and look at the log in detail. |
| 16:27 |
Dyrcona |
Hmm.. maybe not. It's not as different as I thought. |
| 16:27 |
Dyrcona |
Thanks, anyway. |
| 16:29 |
Dyrcona |
Oh, wait! Copying mine down and doing a diff shows that I'm looking for is there! |
| 16:30 |
Dyrcona |
And, it looks like I didn't copy berick's stuff correctly, possibly. |
| 16:30 |
pinesol_green |
News from qatests: Test Failure <http://testing.evergreen-ils.org/~live> |
| 16:30 |
kmlussier |
uh oh |
| 16:31 |
bshum |
Ho hum |
| 16:32 |
bshum |
Guess lsb-release isn't on the test server |
| 16:32 |
kmlussier |
The new Postgres stuff? |
| 16:33 |
bshum |
kmlussier: Yeah, it's dying on the first try to use lsb-release to detect what linux distro the server is running |
| 16:33 |
bshum |
The command isn't found, so that tool isn't installed |
| 16:33 |
kmlussier |
Well, I guess it wouldn't be Bug Squashing week if we didn't have a failed test or two. |
| 16:34 |
kmlussier |
or three or four |
| 16:36 |
Dyrcona |
Hah! |
| 16:36 |
Dyrcona |
I had the right version. I copied it to the wrong place in a rush last week. |
| 16:39 |
Dyrcona |
jeffdavis++ # one more time. If I hadn't looked at your branch, this would have taken me longer to figure out. |
| 16:40 |
jeffdavis |
Glad it helped! |
| 16:41 |
Dyrcona |
lsb-release should be on every server. |
| 16:41 |
bshum |
That's what I would have thought |
| 16:41 |
Dyrcona |
At least, Debian-base distros. |
| 16:41 |
bshum |
If it is installed, maybe we're not invoking it in a way that it understands |
| 16:42 |
Dyrcona |
Might not be in the user's path. |
| 16:42 |
Dyrcona |
Or, yeah, options changed. |
| 16:42 |
Dyrcona |
I hate when they do that. |
| 16:42 |
Dyrcona |
What does the test server run? |
| 16:43 |
bshum |
It's Debian Wheezy |
| 16:43 |
bshum |
I think |
| 16:43 |
bshum |
phasefx would know best |
| 16:44 |
phasefx |
it's wheezy |
| 16:44 |
bshum |
But yeah, it's choking on checking for the lsb_release command and that's passing an empty value to the pgdg apt source, and that results in a 404 error and boom, the rest dies |
| 16:45 |
bshum |
On my debian wheezy test vm, running "lsb_release -cs" gives me "wheezy" in reply, so I'm not sure |
| 16:45 |
Dyrcona |
Same here. |
| 16:45 |
Dyrcona |
Tests could be running in a chroot with limited utilities available. |
| 16:46 |
phasefx |
lsb_release not found |
| 16:47 |
Dyrcona |
phasefx: Is there a chroot? |
| 16:48 |
phasefx |
Dyrcona: yes, coming from package coreutils |
| 16:48 |
Bmagic |
because I gotta be honest, I am a Michael Bolton fan. For my money it doesnt get better than when he sings..... |
| 16:48 |
Dyrcona |
I mean are tests run in a chroot. |
| 16:49 |
phasefx |
Dyrcona: oh, no |
| 16:49 |
phasefx |
apt-get update is failing on that machine too |
| 16:49 |
phasefx |
Err http://apt.postgresql.org -pgdg/main amd64 Packages |
| 16:49 |
phasefx |
404 Not Found [IP: 87.238.57.227 80] |
| 16:49 |
bshum |
phasefx: Yeah it tried to add the apt source for postgres, but it needs to be wheezy-pgdg, not just -pgdg |
| 16:50 |
bshum |
The new makefile uses lsb_release to grab the name of the Debian distro for the add |
| 16:50 |
Dyrcona |
Is /usr/bin in the path? |
| 16:50 |
* Dyrcona |
wonders how lsb_release came to not be installed.... |
| 16:50 |
kmlussier |
Bmagic: Did you mean to say that out loud? |
| 16:51 |
Bmagic |
lol, yeah, hoping someone would get the reference |
| 16:51 |
bshum |
Bmagic: I got it |
| 16:51 |
Dyrcona |
I liked Michael Bolton when he was a commerical Heavy Metal singer.... |
| 16:51 |
phasefx |
apt-get install lsb_release --> E: Unable to locate package lsb_release |
| 16:51 |
bshum |
PC Load letter... |
| 16:51 |
Dyrcona |
:) |
| 16:51 |
kmlussier |
Dyrcona: I must have missed the heavy metal phase. |
| 16:51 |
bshum |
phasefx: The package is lsb-release |
| 16:51 |
phasefx |
doh |
| 16:52 |
bshum |
The command is lsb_release, got me too :) |
| 16:52 |
Dyrcona |
I was shocked when he started singing what he said he wanted to sing. |
| 16:52 |
bshum |
Bmagic: I have people skills, I am good with dealing with people. |
| 16:52 |
Bmagic |
haha. There we go! |
| 16:52 |
* Dyrcona |
mumbles something about a stapler. |
| 16:52 |
Bmagic |
1999 was a good year |
| 16:52 |
phasefx |
definitely wasn't installed: http://testing.evergreen-ils.org/~live/test.32.html |
| 16:53 |
Dyrcona |
1975 was better. |
| 16:53 |
Bmagic |
lol!!!! |
| 16:53 |
Bmagic |
Dyrcona++ |
| 16:53 |
Dyrcona |
phasefx: It should have been. I think everything practically uses it as a build dependency. |
| 16:54 |
bshum |
We could add lsb-release to the spot where we check for wget and ca-certificates in the pre-reqs. Just on the safe side? |
| 16:54 |
bshum |
But that does seem odd. |
| 16:55 |
phasefx |
I could throw it in where I install git-core, in installer_installer.sh |
| 16:56 |
Dyrcona |
phasefx: You shouldn't have to, though. The base system utilities should install it. |
| 16:56 |
Dyrcona |
How do you build the test server? |
| 16:56 |
Dyrcona |
There's a branch in random, isn't there? |
| 16:57 |
phasefx |
http://git.evergreen-ils.org/?p=working/random.git;a=shortlog;h=refs/heads/collab/phasefx/wheezy_installer |
| 16:57 |
Dyrcona |
Yeah. Just found it! |
| 16:57 |
Dyrcona |
:) |
| 16:58 |
|
khuckins__ joined #evergreen |
| 16:58 |
bshum |
That installer script is for after you've already got Debian Wheezy somewhere though. Maybe during the initial deployment there's some different options being selected. The base system utilities weren't installed? |
| 16:59 |
bshum |
"standard system utilities" |
| 16:59 |
phasefx |
it's a pristine-ish wheezy, but built some time ago. VM snapshot |
| 16:59 |
Dyrcona |
Yeah, I'm guessing that standard system utilities were not installed. |
| 16:59 |
Dyrcona |
They should not let you deselect that. |
| 17:00 |
Dyrcona |
You'd only want to skip that in the rarest of circumstances. |
| 17:00 |
bshum |
So maybe we should add lsb-release as a pre-req to avoid this type of problem in the future |
| 17:00 |
bshum |
Even though it's "unlikely" |
| 17:00 |
Dyrcona |
It won't hurt. |
| 17:01 |
bshum |
Simple patch |
| 17:02 |
phasefx |
feel free to patch it, it's collab. I can refresh the vm and fire off another test run in a bit |
| 17:02 |
bshum |
phasefx: Do me a favor and remove the bad apt source entry |
| 17:02 |
Dyrcona |
Only 2 reasons I'd think you'd skip standard system utilities: 1) you're setting up an embedded system that won't need it, and 2) you're gonna do LFS and you're using Debian to bootstrap it. |
| 17:02 |
bshum |
It should be in /etc/apt/sources.list.d |
| 17:03 |
Dyrcona |
#2 is iffy, 'cause you still might need most of that stuff. |
| 17:03 |
phasefx |
ah, that's a bit more involved.. I'll have to look up how to save a new snapshot |
| 17:03 |
phasefx |
is it killing things now or just cosmetic? |
| 17:03 |
Dyrcona |
kvm/qemu? |
| 17:03 |
phasefx |
xen/ganeti I think |
| 17:03 |
Dyrcona |
OK. I can't help in that case. |
| 17:04 |
phasefx |
I don't mind doing it, just not today |
| 17:10 |
bshum |
This should help: http://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/bshum/add-lsb-release-dep |
| 17:10 |
bshum |
Dyrcona: --^ |
| 17:10 |
bshum |
I just tossed it into the same spot we check to make sure wget and ca-certificates are installed. Should be fine there, even though it's not really a pgsql apt source pre-req |
| 17:11 |
bshum |
So we put that in, and make sure the way is clear on the test server, and next run should be slightly happier, we hope. |
| 17:11 |
* bshum |
wanders off |
| 17:19 |
phasefx |
bshum: Dyrcona: the pristine instance has nothing in /etc/apt/sources.list.d/, so I guess we're good |
| 17:29 |
|
khuckins_ joined #evergreen |
| 17:38 |
|
khuckins__ joined #evergreen |
| 18:09 |
Dyrcona |
phasefx | bshum: I pushed a commit to collab/phasefx/wheezy_installer in the random repo to install lsb-release on wheezy and jessie. |
| 18:10 |
phasefx |
Dyrcona++ |
| 18:10 |
Dyrcona |
I figure that is less intrusive than pushing to Evergreen, but we can if it becomes an issue. |
| 18:11 |
Dyrcona |
I should try these scripts sometime. I always used my own and lately have been doing it by hand. |
| 19:13 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 20:11 |
|
kmlussier joined #evergreen |
| 21:04 |
pinesol_green |
[evergreen|Ben Shum] LP#1668257: RTL CSS change for ebook tables - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=68716e6> |
| 22:04 |
pinesol_green |
[evergreen|Kyle Huckins] LP#1511358 Patron Survey Interface - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=e2ee72f> |
