| Time |
Nick |
Message |
| 07:11 |
|
collum joined #evergreen |
| 09:19 |
|
dguarrac joined #evergreen |
| 09:25 |
|
Dyrcona joined #evergreen |
| 09:45 |
|
sandbergja joined #evergreen |
| 10:08 |
Dyrcona |
Ugh. stared make-release and forgot to update the versions in the installation/upgrade docs. |
| 10:09 |
Dyrcona |
No biggie. I'll stop it, reset, and start over. |
| 10:14 |
|
Christineb joined #evergreen |
| 10:15 |
* Dyrcona |
updates the git server in the middle of releases. What could go wrong? |
| 10:17 |
Dyrcona |
And, if bots are training on our Perl code..... |
| 10:17 |
|
kmlussier joined #evergreen |
| 10:35 |
Dyrcona |
Our build instructions still have little errors in them. I mean, if you follow them literally you'll have issues with the README and RELEASE_NOTES documents. |
| 11:07 |
phasefx |
fwiw, I'm working on a proposal that will involve rewriting those. I want one source for documentation and build automation, so the idea is that the automation part should be able to extract code and steps from the asciidoc |
| 11:09 |
phasefx |
err, not build release instructions yet. Just normal installation for a first step |
| 11:49 |
Dyrcona |
Uh huh. Heard of literate programming? Next thing you know, you'll be reinventing TeX... :) |
| 11:50 |
Dyrcona |
Honestly that sound cool. How 'bout a Perl script with POD? Then you can just run the documentation. |
| 11:51 |
phasefx |
devil is in the details :-) and the will to maintain |
| 11:52 |
phasefx |
I think infrastructure as code is the latest buzz phrase for it |
| 11:52 |
phasefx |
or something close |
| 11:52 |
Dyrcona |
:) |
| 11:53 |
Dyrcona |
Too many buzzwords.... Sometimes I think there are too many ways to do the same thing. |
| 11:53 |
Dyrcona |
AI's gonna replace us all.... NOT. |
| 11:55 |
* TimToady |
watches output from Make scroll by. |
| 11:58 |
phasefx |
there was one sci-fi author, maybe Vernor Vinge, I think he had it right, we'll have software acheaoligists digging through layers of code beyond practical understanding :-) |
| 11:59 |
Dyrcona |
Don't need AI for that. We've got systems so complex that no one person can understand it all. |
| 11:59 |
Dyrcona |
Evergreen for one. |
| 12:00 |
phasefx |
right? in that fictional universe (I don't think AI had much of a presence there), those folks were digging up applications to _use_, even if they were black boxes |
| 12:00 |
phasefx |
"found a matter compiler for you. Should work" |
| 12:00 |
Dyrcona |
phasefx: That sounds familiar when you put it that way. Not sure i read it though. |
| 12:01 |
phasefx |
there was another setting that had a sort of wild Matrix/internet where self-replicating code executed freely |
| 12:30 |
Dyrcona |
Our build process is too monolithic. It ought to be easier to run the individual steps in the make_release script for when mistakes happen, which is nearly ever release. |
| 12:30 |
Dyrcona |
s/ever/every/ |
| 12:35 |
|
jihpringle joined #evergreen |
| 12:37 |
* Dyrcona |
feels like John McCarthy: Programming. You're doing it completely wrong. |
| 12:39 |
csharp_ |
@dunno add maybe all of this *waves arms wildly* was a bad idea |
| 12:39 |
pinesol |
csharp_: The operation succeeded. Dunno #98 added. |
| 13:01 |
jeff |
as the saying goes, "teaching sand to think was a mistake" |
| 13:04 |
csharp_ |
jeff++ |
| 13:11 |
csharp_ |
@band add Dee Dupe |
| 13:11 |
pinesol |
csharp_: Band 'Dee Dupe' added to list |
| 13:21 |
|
jihpringle joined #evergreen |
| 13:30 |
csharp_ |
the null authtoken issue I mentioned yesterday is still killing us |
| 13:41 |
jeffdavis |
Are pcrud requests with null authtokens ever valid? |
| 13:44 |
csharp_ |
I don't think so |
| 14:01 |
jeffdavis |
Are you able to tell if the temporary operator change timeout is the source of the requests? |
| 14:19 |
Dyrcona |
I think some pcrud requests do accept a null authtoken, but they are limited. |
| 14:20 |
Dyrcona |
null authtoken usually means your session logged out and the client didn't get the memo. |
| 14:56 |
csharp_ |
seeing attempts to retrieve the session, then open-ils.auth open-ils.auth.session.delete , then the bad calls loop over and over |
| 14:57 |
Dyrcona |
That sounds like the prepatch behavior of the logout spamming. |
| 14:57 |
csharp_ |
hmmm |
| 15:05 |
csharp_ |
open-ils.pcrud open-ils.pcrud.search.pgt null, {"parent":null}, {"flesh":-1,"flesh_fields":{"pgt":["children"]}} - that's the call it's making - trying to see what calls that so we can add a similar check for the token |
| 15:09 |
csharp_ |
this looks like it: https://git.evergreen-ils.org/?p=Evergreen.git;a=blob;f=Open-ILS/web/js/ui/default/staff/circ/patron/bucket/app.js;h=b8d0e8f54448c071bfa466519e898a1dd4f01dbe;hb=HEAD#l633 |
| 15:12 |
csharp_ |
looks like that call is in 5 separate files, including regctl.js and patron_search.js |
| 15:12 |
* csharp_ |
squints hard at patron search |
| 15:20 |
csharp_ |
wondering if we need more checking for egAuth.token() in the pcrud.js service? |
| 15:21 |
csharp_ |
I hate not having a firm grasp on Angular/(JS) - I just don't know the idioms well |
| 15:22 |
csharp_ |
happening right now - thousands of repeated calls from a single client |
| 15:22 |
csharp_ |
wondering how we haven't seen this before |
| 15:23 |
csharp_ |
or maybe we have, but I don't remembering pushing system load |
| 15:23 |
csharp_ |
@monologue |
| 15:23 |
pinesol |
csharp_: Your current monologue is at least 11 lines long. |
| 15:25 |
csharp_ |
and now from 2 clients |
| 15:29 |
|
kmlussier left #evergreen |
| 15:30 |
csharp_ |
restarting services on the affected app server appears to kill off the repeated calls |
| 15:30 |
csharp_ |
but damn |
| 15:33 |
csharp_ |
don't do that, Evergreen, you jerk |
| 15:33 |
csharp_ |
@blame Evergreen |
| 15:33 |
pinesol |
csharp_: Evergreen musta been an Apple employee. |
| 15:43 |
|
jihpringle joined #evergreen |
| 15:45 |
Bmagic |
lol |
| 15:45 |
* Dyrcona |
start over at "Building release dump" |
| 15:46 |
Bmagic |
Dyrcona++ |
| 15:46 |
Dyrcona |
s/start/starts/ |
| 15:46 |
Dyrcona |
@blame Czech |
| 15:46 |
pinesol |
Dyrcona: Czech has a 10 character limit. |
| 15:46 |
Dyrcona |
Hah! |
| 15:46 |
Dyrcona |
@blame Interpolation |
| 15:46 |
pinesol |
Dyrcona: Interpolation caused the white screen of death! |
| 15:47 |
Dyrcona |
Well, there you go.... :) |
| 15:54 |
|
ChurchillComputi joined #evergreen |
| 15:58 |
pinesol |
News from commits: Translation updates - xtb files - main <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=988a668f3342e7e1fa9573e40cd17feb9bb4a286> |
| 16:01 |
Dyrcona |
Bmagic++ sandbergja++ |
| 16:02 |
Bmagic |
back atcha |
| 16:10 |
Dyrcona |
Looks like we don't use ORSF_JS_PATH (-j) any more. It's checked and if it isn't set, the build bails. However, other than checking if it is set, we don't use it. I think it was something needed for the XUL client and we can probably remove it. |
| 16:18 |
csharp_ |
I can't seem to find the right mojo to match the quipu ecard submit path for apache |
| 16:20 |
csharp_ |
might have rubberducked my way into realizing my mistake |
| 16:20 |
csharp_ |
YESSS |
| 16:20 |
Dyrcona |
csharp_: There you go! You're mojo's risin'. |
| 16:21 |
csharp_ |
Mr. MO jo RIS in |
| 16:21 |
Dyrcona |
Ugh... I can spell. I swear it. |
| 16:21 |
Dyrcona |
*your*..... Although you could be the mojo... What do I know? |
| 16:22 |
Dyrcona |
Diggin' in the dirt about an hour ago.... |
| 16:22 |
jeffdavis |
csharp_: if that's related to the notbot stuff, I put a branch in the LP bug that might help with quipu |
| 16:23 |
csharp_ |
<Location /eg/opac/ecard/submit> |
| 16:23 |
csharp_ |
jeffdavis: I'll take a look! |
| 16:24 |
csharp_ |
Bmagic: since you asked yesterday - see my <Location comment |
| 16:24 |
Dyrcona |
Bmagic: FWIW, I got no translation errors when making the tarball this time. |
| 16:24 |
Bmagic |
!! woot |
| 16:25 |
csharp_ |
jeffdavis: nice |
| 16:25 |
csharp_ |
since I have a working solution and in the waning few minutes before I call myself on vacation, I'mma let the apache hack do the job |
| 16:26 |
csharp_ |
I'll review for real when I'm back June 30 |
| 16:26 |
Dyrcona |
csharp_: Enjoy your vacation! |
| 16:26 |
csharp_ |
Dyrcona: thanks! |
| 16:26 |
csharp_ |
much needed |
| 16:27 |
csharp_ |
the notbot fix just dropped my system load by 50% |
| 16:27 |
csharp_ |
the repeated calls probably haven't been the bane of my existence all day |
| 16:31 |
Dyrcona |
The repeated calls are more likely to lead to "no children available" but not so much high load on Apache except as a secondary effect. |
| 16:31 |
csharp_ |
makes sense |
| 16:35 |
Bmagic |
csharp_++ |
| 17:58 |
pinesol |
News from commits: Forward-port 3.14.7 upgrade script <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=aee5289e4334e678cc9649aa0cb51eecc9a51641> |
| 19:28 |
pinesol |
News from commits: Forward port 3.15.1 to 3.15.2 db upgrade script <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=3bc427ad3132c9ddd67a314c8ee596a9ccd2570f> |
| 21:13 |
|
jihpringle joined #evergreen |
