| Time |
Nick |
Message |
| 02:31 |
|
gsams joined #evergreen |
| 07:20 |
|
rjackson_isl joined #evergreen |
| 07:22 |
|
JBoyer joined #evergreen |
| 07:24 |
|
agoben joined #evergreen |
| 07:47 |
|
ericar joined #evergreen |
| 07:47 |
|
ericar joined #evergreen |
| 08:02 |
|
gsams_ joined #evergreen |
| 08:04 |
|
gsams_ joined #evergreen |
| 08:36 |
|
ericar joined #evergreen |
| 08:38 |
|
mmorgan joined #evergreen |
| 08:52 |
|
ericar joined #evergreen |
| 08:53 |
|
Dyrcona joined #evergreen |
| 09:04 |
|
maryj joined #evergreen |
| 09:10 |
|
yboston joined #evergreen |
| 09:22 |
|
ericar joined #evergreen |
| 09:36 |
|
mdriscoll joined #evergreen |
| 09:42 |
bshum |
Whatever happened to the days of eth0 and eth1? enp0s3? Really? |
| 09:43 |
JBoyer |
bshum, if you've ever had eth0 move out from under you unexpectedly that would be a lot more appealing. :) |
| 09:46 |
bshum |
Meh, I suppose |
| 09:48 |
JBoyer |
Done correctly you can also tell which NICs are which in a crowded server without having to play the "carrier, carrier, who's got a carrier?" game. |
| 10:00 |
|
mmorgan1 joined #evergreen |
| 10:01 |
phasefx |
+++ATH0 |
| 10:08 |
JBoyer |
phasefx++ |
| 10:18 |
tsbere |
bshum / JBoyer: My first experience with something like enp0s3 was "where did my previously fully configured and working eth1 go!?!?!?!" >_> |
| 10:20 |
JBoyer |
Well, that's also a possibility and not the best time to find out about that feature. Having never done an in-place upgrade I've not run into that. (Also seems like a missed opportunity for a setup migration thing...) |
| 10:24 |
|
ericar joined #evergreen |
| 10:45 |
tsbere |
JBoyer: It was a very bad time to find out as I had to go visit the server to figure out why it wasn't on the network anymore. Also, the only thing upgraded at that time was the kernel, not the OS version. :( |
| 10:46 |
JBoyer |
Oh, wow. That's a terrible feature to just drop in a regular update. :/ |
| 10:47 |
|
ericar joined #evergreen |
| 10:49 |
tsbere |
JBoyer: Apparently the kernel changed what driver was loading for the network card, which caused it to lose track of the fact the card was previously named eth1, which caused my headaches. :( |
| 10:50 |
JBoyer |
Oh, I thought you meant it was caused by a switch to the permanent names feature. That would be a huge mistake to just drop in a regular update, this sounds like a more subtly annoying thing. |
| 10:53 |
tsbere |
It was a combination of "the kernel supports the new names" and "the kernel decided a different driver was the better choice for the card, so it ignored that it had the old names in a config file" |
| 10:53 |
* tsbere |
ended up ensuring said config file had the old names purged in case a future kernel went back to the old driver, though |
| 10:59 |
pastebot |
"bshum" at 64.57.241.14 pasted "Error, yay? on 'grunt all' for webclient building" (13 lines) at http://paste.evergreen-ils.org/23 |
| 11:06 |
jeff |
bshum: i also encountered that with a branch where i was cherry-picking a bunch of webstaff stuff. did you encounter it on master, or another branch? |
| 11:07 |
bshum |
jeff: Most recent master |
| 11:07 |
bshum |
jeff: So presumably whatever's new is stuff merged from sprint3 things |
| 11:10 |
Dyrcona |
bshum: I haven't seen that one, yet, but I haven't built master since last Tuesday-ish. |
| 11:10 |
Dyrcona |
Want me to verify it? |
| 11:11 |
bshum |
Dyrcona: Sure, if you can get past the node stuff. npm install took *forever* to run successfully for me |
| 11:11 |
bshum |
It would just hang for awhile, but my latest xenial VM grabbed stuff better |
| 11:11 |
|
Christineb joined #evergreen |
| 11:14 |
|
abowling joined #evergreen |
| 11:22 |
jeff |
likely something in bug 1350042 can't handle something that's new in the IDL. |
| 11:22 |
pinesol_green |
Launchpad bug 1350042 in Evergreen "Browser staff client / merge prep" [Wishlist,Fix released] https://launchpad.net/bugs/1350042 |
| 11:24 |
jeff |
first guess is commit 91f844b |
| 11:24 |
jeff |
er. |
| 11:25 |
jeff |
first guess is commit 59af472 |
| 11:25 |
pinesol_green |
jeff: [evergreen|Mike Rylander] Report Templates! - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=59af472> |
| 11:26 |
Dyrcona |
jeff: Don't second guess yourself. :) |
| 11:27 |
jeff |
same guess, two different hashes. ;-) |
| 11:28 |
|
mmorgan joined #evergreen |
| 11:30 |
|
brahmina joined #evergreen |
| 11:36 |
Dyrcona |
Ah, that file gets generated.... |
| 11:36 |
* Dyrcona |
should have remembered. |
| 11:38 |
jeff |
it appears to be the SQL strings. |
| 11:40 |
* Dyrcona |
didn't get that far. ;) |
| 11:52 |
jeff |
Open-ILS/web/js/ui/default/staff/test/data/idl2js.pl outputs an IDL2js.js for testing which is not valid javascript. |
| 11:52 |
jeff |
doesn't seem to be a problem for the OpenILS::WWW::IDL2js handler, though... thus, /IDL2js doesn't seem to break (or generate invalid output) |
| 12:01 |
jeff |
possibly because something in that chain already strips out newlines -- maybe the locale bits in the subrequest that /IDL2js uses |
| 12:04 |
|
jihpringle joined #evergreen |
| 12:17 |
eeevil |
jeff: yeah, I updated the handler to strip newlines (and sql comments, required by the newline stripping). I wasn't actually aware of the test script |
| 12:18 |
jeff |
just got to that part. |
| 12:20 |
jeff |
I believe those two things are the only current (stock) users of fm_IDL2js.xsl |
| 12:30 |
jeff |
bshum: did you open a bug yet? |
| 12:30 |
jeff |
i have a branch. |
| 12:31 |
bshum |
jeff: I haven't had a chance to file a bug on it yet. Been juggling a few things today |
| 12:31 |
jeff |
:-) |
| 12:37 |
|
_adb joined #evergreen |
| 12:37 |
|
ericar joined #evergreen |
| 12:37 |
|
ericar joined #evergreen |
| 12:43 |
|
maryj joined #evergreen |
| 13:01 |
|
dteston joined #evergreen |
| 13:02 |
bshum |
Closed the milestone for 2.11-beta and added a new one for 2.11-rc |
| 13:03 |
jeff |
heh |
| 13:04 |
jeff |
my launchpad hasn't noticed yet. |
| 13:04 |
jeff |
there we go |
| 13:04 |
jeff |
bug 1618136 has my proposed fix |
| 13:04 |
pinesol_green |
Launchpad bug 1618136 in Evergreen "webstaff build/test failures related to IDL2js" [Undecided,In progress] https://launchpad.net/bugs/1618136 - Assigned to Jeff Godin (jgodin) |
| 13:05 |
* jeff |
unassigns self |
| 13:05 |
jeff |
i opted for "preprocess in the same way" as opposed to "try to do those transforms in XSL so that fm_IDL2js.xsl does what the filename claims" :-) |
| 13:06 |
bshum |
Haha |
| 13:06 |
bshum |
I just made a bug too, guess I'll kill mine :) |
| 13:07 |
jeff |
I also considered adding a comment to fm_IDL2js.xsl noting the preprocessing requirement, but didn't want to overcomplicate the fix. |
| 13:08 |
|
Dyrcona joined #evergreen |
| 13:10 |
jeff |
though on the simple-comlicated continuum there are far more complicated options: maybe we need a util package to encapsulate this and all future IDL pre-processing... or, an open-ils.idl_preprocessor service! |
| 13:11 |
* jeff |
stops being silly |
| 13:27 |
* tsbere |
just fought with a Windows server that, after installing updates, decided it was on a public network and thus didn't want anything to do with things connecting to it |
| 13:31 |
jeff |
that sounds... annoyingly similar to your kernel update -> different ethernet driver -> different device name woes from earlier in scrollback. |
| 13:32 |
tsbere |
jeff: Doesn't it? Differences including that it isn't a server I normally maintain, I wasn't installing the updates, and the network was *configured*, it just decided to change firewall modes on it. |
| 13:33 |
tsbere |
jeff: Oh, and the person who was installing updates is supposed to be on vacation and emailed me to go deal with the problem. |
| 13:37 |
bshum |
Tested the patch jeff and it does let me get to the next steps and have a working web client |
| 13:46 |
eeevil |
jeff: I'd be 100% in favor of having the xslt do the processing -- I'm too tuit poor to dig back into xslt string mangling now, though. as evidenced by my "fix" for getting any IDL views to work at all ;) |
| 15:25 |
|
abowling left #evergreen |
| 15:38 |
Dyrcona |
jeff: Do you think the branch on bug 1618136 is ready to go? bshum tried it and I can try it and possibly push it to master tonight. |
| 15:38 |
pinesol_green |
Launchpad bug 1618136 in Evergreen "webstaff build/test failures related to IDL2js" [Undecided,In progress] https://launchpad.net/bugs/1618136 |
| 15:44 |
jeff |
Dyrcona: IMO it's ready to go, yes. at some future point we might return and enhance fm_IDL2js.xsl to no longer require pre-processing, but the branch as it stands fixes the test failure. |
| 15:44 |
Dyrcona |
OK. I can agree with that, but fixing it for now is good enough for now. :) |
| 15:45 |
Dyrcona |
I'll give it a whirl this afternoon/tonight unless someone else beats me to pushing it. |
| 15:46 |
Dyrcona |
I still need to really look at the timezone branch, too. Other things come up, y'know? :( |
| 15:46 |
jeff |
I think it follows with the original spirit of the idl2js.pl support script: transform the IDL to js in a way similar to how OpenILS::WWW::IDL2js does, but do it without requiring that we already have a running system. |
| 15:47 |
jeff |
My branch just updates idl2js.pl to keep up with some new things that OpenILS::WWW::IDL2js does. |
| 15:47 |
|
jvwoolf joined #evergreen |
| 15:48 |
Dyrcona |
Yeah. I eyeballed the diff already. It looks good. I asked bshum to sign off since he tested already. |
| 15:48 |
Dyrcona |
I don't think 3 signoffs would hurt. :) |
| 15:49 |
eeevil |
jeff++ |
| 15:50 |
eeevil |
Dyrcona++ |
| 15:51 |
bshum |
Dyrcona: I added my signoff branch to the bug and updated the status markers |
| 15:51 |
Dyrcona |
bshum++ |
| 16:09 |
Dyrcona |
miker: Do you have a problem with a fix for bug 1618183 going in at this point? It makes Spanish translations available by default. |
| 16:09 |
pinesol_green |
Launchpad bug 1618183 in Evergreen "add Spanish to config.i18n_locale" [Medium,Triaged] https://launchpad.net/bugs/1618183 |
| 16:11 |
miker |
Dyrcona: I don't. i18n is fair game |
| 16:11 |
Dyrcona |
Thanks for confirming, that is what I thought. |
| 16:12 |
Dyrcona |
heh. |
| 16:13 |
* Dyrcona |
just realized that he can't connect to his vm on his laptop while using the staff wireless in the office. |
| 16:14 |
Dyrcona |
The IP address of the wireless card is hardcoded and breaks the connection with the bridge. |
| 16:16 |
jeff |
man, you should dockerize your ansible chef's saltstack or something to prevent that. |
| 16:16 |
Dyrcona |
Yeah... |
| 16:17 |
Dyrcona |
Switching to the public wifi or waiting until I get home is just not complicated enough. :) |
| 16:19 |
jeff |
:-) |
| 16:21 |
Dyrcona |
I should look into Ansible sometime, though. |
| 16:21 |
Dyrcona |
jeff and JBoyer almost have me convinced. :) |
| 16:23 |
Dyrcona |
Another fun thing about the wifi, I ran into bug 1542733 while setting it up. |
| 16:23 |
pinesol_green |
Launchpad bug 1542733 in network-manager (Ubuntu) "Connect to Hidden Wi-Fi Network, "Connect" grayed out" [High,Confirmed] https://launchpad.net/bugs/1542733 |
| 16:24 |
Dyrcona |
Network manager apparently doesn't like connecting to hidden wifi networks with a static IP address. |
| 16:26 |
jeff |
Apple devices have pretty much ended our use of "hidden" SSIDs. |
| 16:27 |
tsbere |
jeff: Oooh, so if I want to annoy Apple users I should hide more of our SSIDs? |
| 16:27 |
* tsbere |
makes note of that ;) |
| 16:35 |
|
maryj joined #evergreen |
| 16:59 |
|
TARA joined #evergreen |
| 17:13 |
|
mmorgan left #evergreen |
| 17:26 |
|
jihpringle joined #evergreen |
| 17:32 |
pinesol_green |
[evergreen|Jeff Godin] LP#1618136 Fix webstaff IDL2js.js test failures - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=6bc0bc2> |
| 17:36 |
|
jvwoolf left #evergreen |
| 17:37 |
dbwells |
grabbing 1000 |
| 17:42 |
pinesol_green |
[evergreen|Ben Shum] LP#1618183: Add Spanish to config.i18n_locale - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=dddb70d> |
| 17:42 |
pinesol_green |
[evergreen|Dan Wells] Stamping upgrade script for Spanish locale seed data - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=9b47089> |
| 18:17 |
|
Dyrcona joined #evergreen |
| 18:18 |
Dyrcona |
Ah, man. My drive home takes too long. ;) |
| 18:18 |
Dyrcona |
dbwells++ |
| 18:20 |
Dyrcona |
Well, I don't think it was just using the wifi at work that prevented my ssh to the vms, 'cause it isn't working now, either. |
| 18:25 |
dbwells |
Dyrcona: If you want something to do, a hot, fresh tarball could use some testing: https://evergreen-ils.org/downloads/Evergreen-ILS-2.11.beta.tar.gz |
| 18:26 |
* dbwells |
heads home |
| 18:27 |
Dyrcona |
OK. I'm trying to figure out why my vm networking is broken. |
| 18:27 |
bshum |
dbwells: One potential reason not to backport the Spanish thing is cause it's an INSERT and if someone upgrades to 2.10.next with it, then try to upgrade to 2.11.0, they'll likely bomb their upgrade due to a rollback when it tries to INSERT the same entry a second time later on |
| 18:28 |
bshum |
Either that or we try to safely add it in a separate part of the 2.11 upgrade script |
| 18:28 |
bshum |
So that it's in its own transaction (probably safer) |
| 18:28 |
Dyrcona |
All right, trusty works. |
| 18:31 |
bshum |
I'll defer to gmcharlt on backport opinions; I'm fine with just focusing on the future with 2.11 |
| 18:34 |
|
berick joined #evergreen |
| 18:34 |
Dyrcona |
So, apparently, I built the xenial vm but didn't fix the networking after. |
| 18:34 |
Dyrcona |
bshum: Did you test the new tarball on xenial? |
| 18:35 |
bshum |
Dyrcona: Not yet, I just got home too :) |
| 18:35 |
Dyrcona |
Well, I'll test trusty if you test xenial. ;) |
| 18:35 |
bshum |
Haha, deal :) |
| 18:35 |
bshum |
I'll spin up a new VM install in my virtualbox |
| 18:35 |
bshum |
I got 25 minutes till dinner anyways :D |
| 18:36 |
bshum |
Actually |
| 18:36 |
bshum |
I guess it doesn't have to be completely fresh |
| 18:36 |
bshum |
I still have my failed Xenial VM from before |
| 18:36 |
bshum |
I can grab the tarball and finish the rest of the Evergreen install setup |
| 18:37 |
bshum |
I stopped at prereq installing earlier |
| 18:37 |
bshum |
So, perfect. |
| 18:37 |
Dyrcona |
Yeah, I was going to install on trusty where I tested the 2.9.7 tarball, but s'pose I could build a fresh one. |
| 18:38 |
Dyrcona |
I should build a vm just with opensrf and save a snapshot. |
| 18:39 |
bshum |
Heh, perhaps |
| 18:39 |
bshum |
I was just thinking the same thing actually |
| 18:39 |
bshum |
Though we should probably gather up what's missing to get OpenSRF 2.5 done |
| 18:43 |
Dyrcona |
Well, you'd have to use OpenSRF master on xenial, wouldn't you? |
| 18:43 |
bshum |
Yes, I would |
| 18:43 |
bshum |
And yes, I did. |
| 18:43 |
bshum |
Huh |
| 18:44 |
bshum |
Oh, wait, nevermind |
| 18:44 |
bshum |
Haha |
| 18:44 |
bshum |
Almost made a mistake :) |
| 18:44 |
Dyrcona |
I'm just going to install over top of 2.9.7, like an upgrade, but build the db fresh. |
| 18:44 |
bshum |
This is what happens when you have both an "Evergreen" dir and "Evergreen-ILS-2.11-beta" dir |
| 18:44 |
bshum |
:D |
| 18:44 |
Dyrcona |
Heh. |
| 18:44 |
bshum |
I went to the wrong place and was like, wait, where are my apache config files?! |
| 18:44 |
bshum |
But I was in the wrong place |
| 18:44 |
Dyrcona |
I just have the tarballs on this vm. |
| 18:48 |
* bshum |
twiddles his thumbs waiting for postgres to setup and load |
| 18:53 |
bshum |
Okay, 2.11 beta installed successfully for me on Xenial |
| 18:53 |
bshum |
Web staff client bits are showing too |
| 18:55 |
bshum |
And spanish and french works for TPAC |
| 18:55 |
bshum |
Looks good to me |
| 18:55 |
bshum |
dbwells++ # quick rebuild |
| 18:56 |
Dyrcona |
Not much in the way of apache config changes from 2.9.7 to 2.11. |
| 18:56 |
Dyrcona |
Just a requiressl |
| 19:03 |
Dyrcona |
opensrf.xml has quite a few changes, but I knew that. :) |
| 19:09 |
Dyrcona |
Oh. That's fortunate. I used a different name for the database, so I can still easily any potential upgrade from 2.9.7 to 2.9.8. :) |
| 19:13 |
Dyrcona |
Well, it woks for me. I tried the OPAC and the browser staff client. |
| 19:14 |
bshum |
Dyrcona++ |
| 19:14 |
Dyrcona |
Aw, shucks...T'weren't nothin'.... ;) |
| 19:15 |
Dyrcona |
bshum++ # For finding the test issue. |
| 19:15 |
Dyrcona |
jgodin++ |
| 19:15 |
Dyrcona |
jeff++ |
| 19:15 |
bshum |
csharp++ # he got me looking for it in the first place |
| 19:15 |
Dyrcona |
For fixing it. |
| 19:18 |
Dyrcona |
That was a pretty easy "upgrade." Wish they were all that simple. :) |
| 19:18 |
Dyrcona |
Of course, you don't normally build a new database for an upgrade, so that made it easier. |
| 19:19 |
* bshum |
wanders off to dinner |
| 19:19 |
Dyrcona |
Yeah, me, too. I believe it is ready. |
| 19:33 |
|
gsams joined #evergreen |
| 19:50 |
|
Shae joined #evergreen |
| 19:58 |
|
dcook joined #evergreen |
| 20:27 |
|
gsams_ joined #evergreen |
| 21:29 |
|
serflog joined #evergreen |
| 21:29 |
|
Topic for #evergreen is now Welcome to the #evergreen library system channel! | We are publicly logged: http://irc.evergreen-ils.org/evergreen | Large pastes at http://paste.evergreen-ils.org |
| 21:42 |
|
TARA joined #evergreen |
| 23:10 |
|
genpaku joined #evergreen |
| 23:16 |
|
genpaku joined #evergreen |
