| Time |
Nick |
Message |
| 00:25 |
|
JBoyer joined #evergreen |
| 02:11 |
|
book` joined #evergreen |
| 05:29 |
|
jgoodson joined #evergreen |
| 05:29 |
|
abneiman joined #evergreen |
| 06:40 |
|
agoben joined #evergreen |
| 07:07 |
|
rjackson_isl joined #evergreen |
| 07:08 |
|
rfrasur joined #evergreen |
| 07:53 |
|
Dyrcona joined #evergreen |
| 08:30 |
|
sandbergja joined #evergreen |
| 08:41 |
|
mmorgan joined #evergreen |
| 09:13 |
* csharp |
says hey to all of y'all |
| 09:14 |
csharp |
@coffee [someone] |
| 09:14 |
* pinesol |
brews and pours a cup of El Salvador La Montana Pacamara, and sends it sliding down the bar to RBecker |
| 09:14 |
csharp |
@tea [someone] |
| 09:14 |
* pinesol |
brews and pours a pot of Golden Orchid, and sends it sliding down the bar to egbuilder (http://ratetea.com/tea/whispering-pines/golden-orchid/7244/) |
| 09:14 |
* csharp |
watches egbuilder emit smoke after having tea splashed on it |
| 09:15 |
|
jvwoolf joined #evergreen |
| 09:16 |
Dyrcona |
@tea csharp |
| 09:16 |
* pinesol |
brews and pours a pot of Honey Black Tea, and sends it sliding down the bar to csharp (http://ratetea.com/tea/health-and-tea/honey-black-tea/7529/) |
| 09:17 |
csharp |
Dyrcona++ |
| 09:17 |
mmorgan |
Good Morning csharp and all! |
| 09:22 |
|
aabbee joined #evergreen |
| 09:26 |
Dyrcona |
"Good morning, good morning, good morning...uh!" |
| 09:45 |
rjackson_isl |
On the bright side: 3 day weekend here. On the not so bright side: 70 and rain with overnight low of 32 |
| 09:49 |
agoben |
Yeah, the rain is not ideal |
| 09:49 |
rjackson_isl |
but sun for the 3 days off with warming temps so yeah!!! |
| 09:50 |
mmorgan |
@love fall weather |
| 09:50 |
pinesol |
mmorgan: The operation succeeded. mmorgan loves fall weather. |
| 10:32 |
mmorgan |
I have a patron record on a training server that I can't save. I click Save, and other than the Save button turning grey, nothing happens. No error in the console or the logs, no pop up message. Anyone seen anything like that? |
| 10:36 |
berick |
mmorgan: i have not. the patron may have super powers. |
| 10:37 |
mmorgan |
Darn those patrons and their super powers! |
| 10:42 |
berick |
mmorgan: do you see a network call in the console? egNet <stuff> <more stuff> |
| 10:45 |
mmorgan |
berick: I see 'egNet open-ils.actor.patron.update' followed quickly by 'closing websocket' |
| 10:46 |
berick |
mmorgan: ok, suggests something dying on the server. to the server logs! |
| 10:49 |
|
jvwoolf joined #evergreen |
| 10:50 |
mmorgan |
osrfsys.log doesn't even show the open-ils.actor.patron.update call |
| 10:50 |
mmorgan |
I can edit other patrons, and can see that call for those edits. |
| 10:51 |
berick |
mmorgan: if you have a separate gateway log, anything there? |
| 10:53 |
berick |
if nothing is there or the nginx/haproxy/apache access logs, then it must be dying on the client before it gets that far |
| 10:54 |
mmorgan |
Ok, I do see it in the gateway log |
| 10:56 |
mmorgan |
Received <stream:error> message from Jabber server |
| 10:56 |
mmorgan |
Text of error message received from Jabber: XML stanza is too big |
| 10:56 |
berick |
ding ding ding |
| 10:57 |
miker |
dang, they must have a really long name! ;) |
| 11:00 |
mmorgan |
Nothing looks unusual about the patron record. Any other log to look at for more information? |
| 11:01 |
berick |
Bobby Tables cousin, Little Bobby Zillion Zero-Width Spaces |
| 11:01 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 11:04 |
Dyrcona |
mmorgan: Increase max_stanza_size in ejabberd.{yml,cfg} |
| 11:08 |
berick |
mmorgan: it's very odd. anything look unusual in the edit form? super long values, maybe stat cats?. I'd also check the rows in actor.usr_setting for that user, could be some chaos there. |
| 11:12 |
Dyrcona |
Is this a staff account with lots of copy templates, perchance? |
| 11:13 |
berick |
Dyrcona: kind of what I'm thinking |
| 11:13 |
mmorgan |
Nope, patron account, no templates. |
| 11:15 |
mmorgan |
The patron has a stat cat entry, I can see it in the db, but don |
| 11:16 |
mmorgan |
't see it in the edit screen |
| 11:19 |
rjackson_isl |
mmorgan are you using a workstation registered to same home ou as patron (thinking about stat cats matching up) |
| 11:21 |
mmorgan |
rjackson_isl: I was not, but logged in to a workstation that matched the owner of the stat cat and still did not see it. |
| 11:22 |
* mmorgan |
removed the row from stat_cat_entry_usr_map, but still can't save the patron record :-( |
| 11:23 |
Dyrcona |
mmorgan: Yeah. I don't think that's it. This patron has a lot of something that makes the update message too large to go through ejabberd. |
| 11:24 |
rjackson_isl |
address fields in db seem sane? |
| 11:24 |
Dyrcona |
I'm seeing some things that raise more questions than answers: Does user edit really go through XUL or is that just permissions editing? |
| 11:24 |
Dyrcona |
I'm guessing lots of bills, circs, alerts, or notes. |
| 11:27 |
mmorgan |
One address, looks sane. Patron does have bills, penalties, but would those come into play when updating the patron record? |
| 11:28 |
Dyrcona |
Everything gets pulled and if fleshed update is used, everything could get sent back to the server, unless flattened/stripped. |
| 11:29 |
Dyrcona |
I know it's possible, I just don't know exactly what the web staff client does. |
| 11:29 |
jeff |
this is where looking at the websockets data sent from the client could be helpful. it can be a little annoying/tricky to get the data out and pretty print it with jq or something. |
| 11:29 |
mmorgan |
Aha! |
| 11:30 |
jeff |
(this is assuming that the data from the client is what's at issue) |
| 11:30 |
mmorgan |
The patron had a public note, about a paragraph long. I deleted that note, and was then able to save the record. |
| 11:31 |
* mmorgan |
doesn't understand why that would be an issue, but feels better having that info. |
| 11:31 |
berick |
huh |
| 11:31 |
Dyrcona |
mmorgan: You've found a place where chunking/bundling needs more attention. |
| 11:32 |
Dyrcona |
Apparently.... |
| 11:33 |
mmorgan |
Just to confirm, I reentered the note, and again cannot save the record. |
| 11:34 |
Dyrcona |
Right. You pasted "Text of error message received from Jabber: XML stanza is too big" earlier. |
| 11:35 |
Dyrcona |
That means the update message is too big for your ejabberd settings. You can get around that by increasing max_stanza_size in the ejabberd configuration. |
| 11:38 |
Dyrcona |
It also looks like some part of this needs to have chunking/bundling fixes, but I'll leave that to the experts to figure out. :) |
| 11:44 |
rfrasur |
http://www.trbimg.com/img-56c769f8/turbine/bal-punkin-chunkin-may-be-coming-to-maryland-20160219 |
| 11:44 |
berick |
haha we def. should have called it chunkin' / bundlin' |
| 11:45 |
rfrasur |
https://blog.lib.uiowa.edu/eng/files/2016/09/PunkinChunkin_FCC_PeterDutton.jpg |
| 11:45 |
rfrasur |
a bundle of chunkin' |
| 11:45 |
mmorgan |
Heh. |
| 11:45 |
rfrasur |
er chuckin'. whatever. |
| 11:47 |
berick |
fetchez la vache |
| 11:48 |
* rfrasur |
chuckles (related to chunkin' and chuckin') |
| 11:48 |
mmorgan |
:) |
| 11:49 |
rfrasur |
is there a technical term for the "thinking wheel?" like "the hoping to find" search spinner or something? |
| 11:51 |
mmorgan |
FWIW, I just loaded up another patron with a number of large public notes, and didn't have any trouble editing and saving that patron :-/ |
| 11:51 |
rfrasur |
(oh good grief - it's called a "throbber") |
| 11:53 |
* mmorgan |
's head feels that right now! |
| 11:53 |
|
jihpringle joined #evergreen |
| 11:53 |
rfrasur |
mmorgan - you've done good work. It sounds like that patron really might be magic. ish. |
| 11:55 |
|
bos20k joined #evergreen |
| 11:56 |
mmorgan |
Fortunately, this is a test patron, but some real patrons can have records that are just as, err, busy, if not busier! |
| 11:57 |
* mmorgan |
would like to open a lp bug, but is not sure what to report. |
| 11:57 |
berick |
mmorgan: Dyrcona's suggestion of raising the ejabberd max stanza size is the fix you need for now |
| 11:57 |
berick |
and there's already an LP re: the pendin' chunkin' / bundlin' work |
| 11:59 |
* mmorgan |
will pass that on to our systems manager. |
| 11:59 |
mmorgan |
Do we need new lp tags for chunkin' and bundlin' ? ;-) |
| 12:00 |
mmorgan |
berick++ |
| 12:00 |
mmorgan |
Dyrcona++ |
| 12:07 |
berick |
mmorgan++ |
| 12:11 |
Dyrcona |
mmorgan++ |
| 12:12 |
Dyrcona |
I don't think we need tags for chunking and bundling. |
| 12:13 |
mmorgan |
What about punkin' ? ;-) |
| 12:14 |
* miker |
reads up |
| 12:15 |
miker |
what we really need (and I have a perl-only-so-far WIP branch...) is streaming of large params (similar to chunking in results, uses some of the receive-side code) |
| 12:15 |
* miker |
wonders if this might be a good hack-a-way topic |
| 12:20 |
* miker |
adds to discussion topic list |
| 12:22 |
Dyrcona |
streaming via websockets you mean? |
| 12:25 |
mmorgan |
So does bad chunking/bundling generally cause the type of failure where the user clicks and nothing happens? Or could it manifest in other ways? |
| 12:30 |
|
remingtron joined #evergreen |
| 12:39 |
Dyrcona |
mmorgan: Somebody clicking and nothing happening could be anything. Chunking and bundling is meant to get around xml_stanza_size and the XML message to big errors. |
| 12:41 |
mmorgan |
Ok, thanks. That type of failure is particularly annoying. |
| 13:09 |
|
khuckins joined #evergreen |
| 13:22 |
miker |
Dyrcona: the idea is to look at the params and, if too big, send a new "here's the method, but you'll get some followup messages with the params" message |
| 13:24 |
miker |
mmorgan: right now, chunking is only applied to results, not requests. strictly speaking, bundling is different, it packages up multiple responses (from a single streaming response method) in one stanza so we're sending overall fewer xmpp messages (and doing less en/de-coding of JSON and XML) |
| 13:24 |
miker |
bundling provides a good speedup for "send me 1000 tiny objects" calls (as much as .atomic, really) |
| 13:27 |
miker |
(or, nearly as good as .atomic ... because those end up getting chunked :) ) |
| 13:31 |
rfrasur |
microscopic pumpkin https://i.pinimg.com/originals/4e/81/5f/4e815fc555d577070d2273c47c65ca05.jpg |
| 13:41 |
berick |
sequel to The Great Pumpkin |
| 13:41 |
rfrasur |
lol, a very short sequel |
| 13:45 |
rfrasur |
Meanwhile, in the course of reading... https://xkcd.com/927/ |
| 13:47 |
Dyrcona |
rfrasur: Yes, that xkcd is a classic. One of my favorites. |
| 13:47 |
rfrasur |
Dyrcona: a gift that keeps on giving...because it's still so relevant. |
| 13:48 |
Dyrcona |
Yes, always relevant. |
| 14:18 |
jonadab |
Related to "those who do not know about [old technology] are doomed to reimplement it, badly". Which explains *so much* software, and not a few protocols. |
| 14:25 |
Dyrcona |
Kind of like other programming languages and Lisp. :p |
| 14:27 |
pinesol |
[evergreen|Jeff Davis] LP#1842763: Vandelay: disable grid row select on queued record matches - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=cf4703e> |
| 14:28 |
|
jvwoolf joined #evergreen |
| 14:31 |
rhamby |
rfrasur: that comic is a lie. the number of new standard created in such as situation is participants + 1 so it would be at least 16 |
| 14:32 |
Dyrcona |
rhmaby: You're talking about Z39.XX standards where the number of interpretations is infinite. |
| 14:33 |
Dyrcona |
rhamby^^ typos-- |
| 14:33 |
Dyrcona |
My day has actually been typorific. |
| 14:33 |
rhamby |
Dyrcona: I share that experience |
| 14:34 |
Dyrcona |
I have 3 commits in a row in a local scripts reposity: Fix typo.... |
| 14:34 |
berick |
rhamby: for every set of standards, there's an emergent standard that's a mashup of the rest? |
| 14:34 |
Dyrcona |
heh... "reposity" :) I can spell, I swear. I just can't type today. |
| 14:35 |
csharp |
@band add Reposity |
| 14:35 |
pinesol |
csharp: Band 'Reposity' added to list |
| 14:35 |
rhamby |
berick: deconstructionist standards - there is something on paper but you get to decide what it _really_ means,for when english lit majors need real jobs |
| 14:35 |
* rhamby |
was an English Lit major in undergrad |
| 14:36 |
berick |
rhamby: i just read your comment 12 different ways |
| 14:36 |
* csharp |
is also in the English major turned tech crew |
| 14:37 |
rfrasur |
rhamby++ |
| 14:38 |
rhamby |
heh |
| 14:39 |
rfrasur |
That was for the + 1 = 16, fwiw |
| 14:42 |
mmorgan |
spellcheck-- #attempted order, got ogre |
| 14:45 |
|
jvwoolf1 joined #evergreen |
| 14:50 |
|
sandbergja joined #evergreen |
| 17:08 |
|
mmorgan left #evergreen |
| 17:19 |
|
stephengwills joined #evergreen |
| 17:49 |
|
stephengwills joined #evergreen |
| 18:52 |
|
stephengwills joined #evergreen |
| 20:06 |
|
stephengwills joined #evergreen |
| 21:13 |
|
stephengwills joined #evergreen |
| 21:46 |
|
stephengwills joined #evergreen |
| 22:20 |
|
stephengwills joined #evergreen |
| 23:31 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 23:42 |
|
sandbergja joined #evergreen |
| 23:56 |
|
jvwoolf joined #evergreen |
