| Time |
Nick |
Message |
| 03:05 |
|
beanjammin joined #evergreen |
| 03:56 |
|
paxed joined #evergreen |
| 04:01 |
paxed |
gmcharlt: hey, a Q about MARC::Record - i can't seem to make insert_fields_ordered put the new fields in the correct place. eg. https://bilious.alt.org/~paxed/vaara_update_cd.pl.txt puts 650 right after leader? |
| 06:32 |
pinesol_green |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 06:54 |
|
JBoyer joined #evergreen |
| 07:11 |
|
jvwoolf joined #evergreen |
| 07:16 |
|
rjackson_isl joined #evergreen |
| 07:20 |
|
jvwoolf1 joined #evergreen |
| 07:25 |
|
agoben joined #evergreen |
| 08:13 |
|
collum joined #evergreen |
| 08:21 |
csharp |
any plans to release OpenSRF 3.0.1 containing fixes from bug 1744158 and bug 1746577? |
| 08:21 |
pinesol_green |
Launchpad bug 1744158 in OpenSRF "osrf_websocket_translator send requests to the bit-bucket" [Undecided,Fix committed] https://launchpad.net/bugs/1744158 |
| 08:21 |
pinesol_green |
Launchpad bug 1746577 in OpenSRF "Websocket translator responder thread loops on broken jabber socket" [Undecided,Fix committed] https://launchpad.net/bugs/1746577 |
| 08:22 |
csharp |
(our current build process relies on community-released tarballs - we can change that, but if there's something in the works, we can wait for that) |
| 08:23 |
csharp |
and btw, happy to help that happen :-) |
| 08:42 |
|
Dyrcona joined #evergreen |
| 08:43 |
|
mmorgan joined #evergreen |
| 08:46 |
dbwells |
paxed: Good to see you! |
| 08:52 |
paxed |
dbwells: thanks. working with koha at least a bit |
| 08:52 |
* csharp |
waves at paxed |
| 08:52 |
dbwells |
paxed: Do you have an example record where your code is failing, or should it fail for any MARC record? |
| 08:54 |
|
bos20k joined #evergreen |
| 08:59 |
paxed |
dbwells: i'll look one up tomorrow |
| 09:04 |
|
mmorgan1 joined #evergreen |
| 09:04 |
* Dyrcona |
waves to paxed |
| 09:07 |
dbwells |
paxed: I ran a lightly hacked up version of your check_marc() sub on a record here, and 'lautapelit' showed up in the right spot, FWIW. The code inserts before the first field "higher" than the one you have, so I'd be suspicious of your first field, though that's just a place to start. |
| 09:07 |
dbwells |
Well, I guess "higher or equal to". |
| 09:10 |
|
kmlussier joined #evergreen |
| 09:33 |
|
stephengwills joined #evergreen |
| 09:36 |
|
yboston joined #evergreen |
| 09:50 |
csharp |
sandbergja++ # web client dev docs |
| 09:51 |
* csharp |
is catching up from a week off |
| 10:02 |
|
mmorgan1 joined #evergreen |
| 10:02 |
|
krvmga joined #evergreen |
| 10:04 |
|
jvwoolf joined #evergreen |
| 10:49 |
|
Christineb joined #evergreen |
| 10:54 |
pinesol_green |
[evergreen|Dan Wells] Forward-port 3.1.0 upgrade script - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=00af8a0> |
| 11:00 |
|
jvwoolf joined #evergreen |
| 11:18 |
|
beanjammin joined #evergreen |
| 11:30 |
Bmagic |
@coffee [someone] |
| 11:30 |
* pinesol_green |
brews and pours a cup of Colombia El Castillo Micro-Lot Nelson Melo, and sends it sliding down the bar to rjackson_isl |
| 11:30 |
rjackson_isl |
nice - I needed taht Bmagic! |
| 11:31 |
Bmagic |
rjackson_isl: don't we all! |
| 11:31 |
rjackson_isl |
but it still didn't fix my typing woes |
| 11:31 |
Bmagic |
The only solution is: MOAR COFFEE |
| 11:31 |
rjackson_isl |
HA! |
| 11:55 |
|
mmorgan joined #evergreen |
| 11:57 |
|
ngf42 joined #evergreen |
| 11:57 |
|
kmlussier joined #evergreen |
| 12:00 |
|
beanjammin joined #evergreen |
| 12:02 |
Bmagic |
Has this happened to anyone else? |
| 12:02 |
Bmagic |
ERROR: index "ahr_capture_time_current_copy" contains unexpected zero page at block 34098 |
| 12:02 |
kmlussier |
Somebody here was getting a blank screen after logging into the web client. When we looked in the Console, we saw one of those offline db connection error messages, similar to the ones we saw with bug 1718301. |
| 12:02 |
pinesol_green |
Launchpad bug 1718301 in Evergreen "Webstaff offline DB connection failures" [High,Fix released] https://launchpad.net/bugs/1718301 |
| 12:02 |
kmlussier |
The error, in this case, was Constraint error: (201) Duplicate keys are not allowed, index: Setting.pkSetting, key: "circ.staff_client.receipt.notice_text" |
| 12:03 |
kmlussier |
I don't have any trouble on the same server. |
| 12:05 |
* kmlussier |
can file a bug, but wanted to make sure it is a bug and not something else. |
| 12:06 |
|
khuckins joined #evergreen |
| 12:11 |
JBoyer |
kmlussier, The trouble with those bugs is that I've never found any steps to cause it. I know we've seen it after 3.0 but I can't be much more helpful than that. :/ |
| 12:13 |
JBoyer |
Bmagic, I can't say I've ever come across that myself. At least in that particular case it's not hard to drop and rebuild that index but it certainly seems like you've run into an alarming amount of FS-level corruption in your database. Everything ok day-to-day? |
| 12:15 |
berick |
weird, i have hatch running on a windows 10 machine. works fine, but I can't find the NativeMessagingHosts registry key entries |
| 12:15 |
berick |
not under HKEY_LOCAL_MACHINE or HKEY_CURRENT_USER |
| 12:15 |
berick |
or anywhere else for that matter |
| 12:15 |
* berick |
is trying to offere suggestions for bug 1756084 |
| 12:15 |
pinesol_green |
Launchpad bug 1756084 in Evergreen "Printer List Empty In Hatch Printer Settings" [Undecided,Confirmed] https://launchpad.net/bugs/1756084 |
| 12:15 |
JBoyer |
berick, looked under the WOW64 subkey? *shouldn't* be there, but I think Chrome will allow it. |
| 12:16 |
berick |
JBoyer: looking.. what would that be under? |
| 12:16 |
JBoyer |
Ah, have to check. |
| 12:17 |
berick |
oh |
| 12:17 |
berick |
i found it |
| 12:17 |
berick |
child of Software |
| 12:17 |
berick |
and sure enough, the hatch stuff is in there |
| 12:17 |
berick |
JBoyer++ |
| 12:17 |
csharp |
@band add Child of Software |
| 12:17 |
pinesol_green |
csharp: Band 'Child of Software' added to list |
| 12:18 |
berick |
WOW6432Node |
| 12:18 |
JBoyer |
That maybe shouldn't have a Children of the Corn feeling to it, but... |
| 12:18 |
JBoyer |
Huh. Forgot it was such an ugly key. Not sure why it's in there though, 32-bit Chrome? |
| 12:18 |
JBoyer |
Though the installer doesn't check for that, so... Hm. |
| 12:19 |
berick |
64-bit chrome / win 10 |
| 12:22 |
JBoyer |
Huh. We never call SetRegView except to check Java versions, so it's always going to be in WOW6432Node. I thought we were using the 'native' hive when possible? |
| 12:23 |
berick |
don't know. FWIW, i've looked in earlier versions of Hatch on older versions of Windows and it was not in the WOW node. |
| 12:25 |
|
khuckins_ joined #evergreen |
| 12:27 |
JBoyer |
I just went through every version in master and the original version used the 64 bit view exclusively, that was lost when I restructured everything. Looking at how things work I think this is the "right" way to do it, but I had to reacquaint myself with some stuff. |
| 12:27 |
|
kdt8 joined #evergreen |
| 12:28 |
JBoyer |
Firefox has some goofy rules about the keys it will look at, but I don't think Chrome actually cares about 32/64. |
| 12:36 |
* csharp |
answers his own question about OpenSRF 3.0.1 by reading the dev meeting minutes |
| 13:03 |
|
kmlussier joined #evergreen |
| 13:08 |
|
jihpringle joined #evergreen |
| 13:08 |
miker |
Bmagic: I'd recommend 1) an immediate reindex (or: create index concurrently, then drop old) 2) check your kernel logs for IO errors. are you using thin provisioned storage for the db? (if so, don't ... it's as reliable as NFS, for many of the same reasons that NFS is unreliable) |
| 13:13 |
Bmagic |
miker: No IO errors (thank goodness). Postgres is on a local disk. It looks like it's just this index which looks like it's custom (not stock eg.) |
| 13:20 |
csharp |
create index concurrently works well (though it's slow on a busy system) |
| 13:22 |
miker |
Bmagic: note, vmware local storage can still be thin provisioned ... if you're using ESX, say. just a heads up |
| 13:23 |
Bmagic |
miker: no worries :) - it's not setup like that |
| 13:38 |
JBoyer |
Bmagic, this is more just for my own curiosity but is your data dir on a RAID (what type?) or a single large storage allocation? |
| 13:40 |
Bmagic |
JBoyer: Amazon offers several options. I have chosen two different options. Both are local SSD, physically located on the same hardware with the VM. In one case, I have two 1900GB SSD's which I have configured in RAID 0 - and the other is a single 480GB SSD non RAID config. Both had the same index issue. |
| 13:41 |
JBoyer |
Bmagic++ |
| 13:42 |
Dyrcona |
Bmagic: You should probably mirror rather than stripe your RAID. |
| 13:42 |
Dyrcona |
But what do I know? |
| 13:42 |
Bmagic |
Dyrcona: I knew that recommendation would be coming next |
| 13:43 |
Dyrcona |
SSDs are generally fast enough. |
| 13:43 |
Bmagic |
The reason I chose not to mirror - Amazon deletes the disks anyway (upon shutdown) - they are released back into the pool |
| 13:43 |
Dyrcona |
Another Q: Is this software RAID or is a controller involved? |
| 13:43 |
Bmagic |
software raid |
| 13:43 |
Dyrcona |
Oh... So these SSDs are not permanently assigned to you. |
| 13:44 |
Bmagic |
They are so long as the machine stays on |
| 13:44 |
* Dyrcona |
adds another reason not to run Evergreen in the "cloud" to his list of reasons not to run Evergreen in the "cloud." :) |
| 13:44 |
Bmagic |
:) |
| 13:44 |
Bmagic |
gotta run |
| 13:45 |
csharp |
wow that would make me super nervous - I know you've explained that to me before, Bmagic, but I get nervous enough about data loss with our full control from bare metal on up :-) |
| 13:45 |
Dyrcona |
Nothing beats ZFS on NVMe. :) |
| 13:49 |
Dyrcona |
For the logs on my Q about software vs. hardware RAID: I'd recommend software RAID with SSDs. That's based on my experience and some discussion from the postgres performance mailing list from a couple of years ago. |
| 13:49 |
Dyrcona |
With modern SSDs, the controller just gets in the way. |
| 14:16 |
|
krvmga joined #evergreen |
| 14:20 |
rjackson_isl |
have staff at site reporting error of 1532:MONEY_BILLABLE_TRANSACTION_SUMMARY_NOT_FOUND and indicating evey time they accept money and apply a payment for copies they have to close the screen and reopen to get it to accept payments |
| 14:21 |
rjackson_isl |
anyone hear/see anything similar? |
| 14:23 |
jeff |
Bmagic: What is the definition of your ahr_capture_time_current_copy index? How are you creating/restoring it on each of the instances? What instance types are you working with? Guessing from clues above, your m2.xlarge with 1 x 420 GB SSD ephemeral instance storage is going to have less-than-optimal performance unless you init/pre-allocate it by writing zeros. The larger 2 x 1900 GB SSD instance |
| 14:23 |
jeff |
shouldn't need that, but one of those types doesn't support TRIM... |
| 14:33 |
kmlussier |
khuckins: Are you still working on bug 1511742? Or were you waiting for some feedback to your question in comment #34? |
| 14:33 |
pinesol_green |
Launchpad bug 1511742 in Evergreen "webclient: Need ability to type to selection in in some menus" [High,Confirmed] https://launchpad.net/bugs/1511742 |
| 14:37 |
khuckins |
kmlussier: Anyone can take that one - at one point I was waiting on feedback from comment #34 but I'd been running into trouble actually getting my ideas set forth in that comment working as I'd wanted. I'm not actively working on that card |
| 14:50 |
kmlussier |
khuckins: OK, thanks! |
| 14:58 |
Dyrcona |
Anyone know how to use the console to inpect an Angular $scope variable? |
| 15:02 |
jeff |
Dyrcona: in Chrome? inspect the relevant element, enter $scope in the console? |
| 15:16 |
JBoyer |
Dyrcona, Sounds like the sort of thing that will require setting a breakpoint in the debugger. |
| 15:17 |
Dyrcona |
I keep getting this: VM882:1 Uncaught ReferenceError: $scope is not defined |
| 15:17 |
Dyrcona |
at <anonymous>:1:1 |
| 15:17 |
* csharp |
would love a tutorial on the best ways to use the dev tools in Chrome/FF |
| 15:17 |
jeff |
Dyrcona: yeah, I'm trying to remember why I thought that worked. Sorry. Might have led you astray. |
| 15:18 |
csharp |
I can intuit a lot and I have Google, but maybe we can do some technique swapping at the conference |
| 15:18 |
Dyrcona |
S'allright. I only tried it once or 3 times. I've been doing other things, too. |
| 15:25 |
JBoyer |
Ick, though. Best you can hope for in a situation like that is something useful in the stack trace. Doing some custom work or testing master? |
| 15:27 |
Dyrcona |
Trying to figure out why item status only wants to delete 1 item instead of all the selected ones. |
| 15:28 |
khuckins |
Dyrcona: http://ng-inspector.org/ is a really handy chrome/firefox/safari extension |
| 15:31 |
Dyrcona |
khuckins: Thanks. I'll take a look. |
| 15:34 |
khuckins |
D'oh that's the wrong link, the better one is ng-inspect, easier to parse: https://chrome.google.com/webstore/detail/ng-inspect-for-angularjs/cidepfmbgngpdapgncfhpecbdhmnnemf?hl=en - for chrome only |
| 15:34 |
khuckins |
Similar names |
| 15:56 |
|
beanjammin joined #evergreen |
| 16:42 |
csharp |
@band add Zuckerberg's Big Mistake |
| 16:42 |
pinesol_green |
csharp: Band 'Zuckerberg's Big Mistake' added to list |
| 16:42 |
* berick |
chuckles |
| 16:51 |
collum |
I love their rendition of "If I could turn back time" |
| 17:01 |
csharp |
collum++ |
| 17:03 |
|
mmorgan left #evergreen |
| 17:10 |
|
jvwoolf left #evergreen |
| 18:30 |
pinesol_green |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 18:42 |
|
abowling joined #evergreen |
| 23:49 |
|
abowling1 joined #evergreen |
