| Time |
Nick |
Message |
| 06:40 |
|
rlefaive joined #evergreen |
| 07:16 |
|
agoben joined #evergreen |
| 07:18 |
|
rjackson_isl joined #evergreen |
| 07:21 |
|
Callender joined #evergreen |
| 07:37 |
|
JBoyer joined #evergreen |
| 08:55 |
|
bos20k joined #evergreen |
| 09:00 |
|
aubjordan joined #evergreen |
| 09:08 |
|
yboston joined #evergreen |
| 09:21 |
|
maryj joined #evergreen |
| 09:36 |
|
Dyrcona joined #evergreen |
| 10:06 |
|
kmlussier joined #evergreen |
| 10:06 |
kmlussier |
Good morning #evergreen! |
| 10:09 |
bshum |
Good morning! :D |
| 10:09 |
|
jvwoolf joined #evergreen |
| 10:10 |
Dyrcona |
Good morning! |
| 10:15 |
csharp |
Good Morning Good Morning Good Morning-uh |
| 10:17 |
Dyrcona |
I've got nothing to say, but it's OK. |
| 10:20 |
|
akilsdonk joined #evergreen |
| 10:31 |
Bmagic_ |
@coffee |
| 10:31 |
* pinesol_green |
brews and pours a cup of Decaf Organic Caffe Volcan, and sends it sliding down the bar to Bmagic_ |
| 10:32 |
|
Bmagic joined #evergreen |
| 10:32 |
kmlussier |
Decaf! |
| 10:33 |
|
bos20k joined #evergreen |
| 10:33 |
csharp |
decaf++ |
| 10:38 |
Bmagic |
Volcan++ |
| 10:51 |
kmlussier |
Not to be confused with Vulcan. |
| 10:51 |
Dyrcona |
Or Vulkan. |
| 10:52 |
csharp |
free_association++ |
| 10:55 |
Dyrcona |
https://en.wikipedia.org/wiki/Vulkan_(API) |
| 11:14 |
Bmagic |
Anyone know what time on November 2nd that things will get started on the hack-a-away? And what time on the 4th things will be wrapped up? The office manager is playing with flight times. |
| 11:19 |
rhamby |
Assuming that we can use the last couple of years as a template we've gotten started at 9 on the first day and informally wrapped up about noon on the last day. |
| 11:27 |
|
_adb joined #evergreen |
| 11:37 |
|
bmills joined #evergreen |
| 11:58 |
|
brahmina joined #evergreen |
| 11:59 |
|
alynn26 joined #evergreen |
| 12:01 |
Bmagic |
rhamby: thanks! |
| 12:03 |
|
jihpringle joined #evergreen |
| 12:06 |
kmlussier |
When working on a stock Evergreen install with Concerto data, if I scope a search to Example sub-library 1, I retrieve results that are owned by its parent (BR1), but are not owned by the sub-library. Is that expected behavior? |
| 12:08 |
bshum |
kmlussier: Are you logged in as anyone or have a preferred library set somehow? |
| 12:08 |
kmlussier |
bshum: No, there's no preferred library. |
| 12:09 |
kmlussier |
http://mlnc4.noblenet.org/eg/opac/results?query=concerto&qtype=keyword&fi%3Asearch_format=&locg=8&detail_record_view=0&sort= |
| 12:10 |
tsbere |
kmlussier: So, in looking at that link I don't see any problems on page one |
| 12:11 |
* tsbere |
also doesn't see any results on page one |
| 12:11 |
kmlussier |
tsbere: Do you see any results that are owned by SL1? |
| 12:11 |
jeff |
i suspect the matter is complicated by the fact that there are physical holdings and located URIs on the same bibs. |
| 12:12 |
kmlussier |
Oh! It's the Located URIs. OK, that makes sense. |
| 12:12 |
tsbere |
kmlussier: Ok, I found out why I don't (didn't) see any results and fixed that, and now I do see results, but like jeff mentioned they all have URIs |
| 12:13 |
* kmlussier |
goes back to investigating the real bug that led me down this path. |
| 12:14 |
* tsbere |
removes all url-rewriting rules related to "mlnc[1-4].*.org" server names in a firefox addon to prevent issues with links >_> |
| 13:04 |
|
gsams joined #evergreen |
| 13:26 |
|
sarabee joined #evergreen |
| 14:06 |
dbs |
jeffdavis noted that they're running yaz4 on ubuntu 14.04 & a z39.50 server with no problems; turns out that their SRU targets are HTTP, not HTTPS |
| 14:07 |
dbs |
Whereas ours are HTTPS, which would explain why libgnutls is identified as the segfault |
| 14:08 |
Dyrcona |
dbs: I think the same is true at MVLC, but it has been a couple of months since I moved the ZURL or worked at MVLC for that matter. ;) |
| 14:08 |
Dyrcona |
So, yours looks like a different, but similar, issue. |
| 14:12 |
* Dyrcona |
remembers why we used to install all of the important stuff from source code at the University of Kentucky, College of Engineering Computing Center.... |
| 14:13 |
Dyrcona |
We'd install Debian server, then if it was going to be a webserver, install Apache, Perl, and the other necessary packages from tarballs. |
| 14:13 |
Dyrcona |
Not so great for long term maintainability, but pretty good for short term reliability. :) |
| 14:14 |
* Dyrcona |
returns to pondering duplicate e-resource records. |
| 14:14 |
dbs |
heh, yep. reminds me of running gentoo for a number of years. made the mistake of running it on a production server though |
| 14:14 |
Dyrcona |
:) |
| 14:15 |
* dbs |
is glad Dyrcona is still an active part of the evergreen community |
| 14:15 |
Dyrcona |
One of the reasons that I like FreeBSD and OpenBSD is ports. Very often up to date code and little breakage, 'cause it was compiled on your machine. |
| 14:15 |
Dyrcona |
Yeah, I work for C/W MARS, now. |
| 14:15 |
Dyrcona |
I never made a big announcement... At least, I don't think I ever published the blog post. |
| 14:16 |
dbs |
yeah, that's the same sort of model Gentoo uses. it's only when glibc or something major got updated that things got super-shakey |
| 14:16 |
Dyrcona |
Yeah.... FreeBSD has a couple of utilities for managing that. |
| 14:17 |
Dyrcona |
I use portupgrade, and the UPDATING file will warn you about "You have to rebuild all the dependent packages..." type of issues. |
| 14:17 |
Dyrcona |
OpenBSD, not so much... |
| 14:18 |
* Dyrcona |
really goes back to the duplicate e-resource records. ;) |
| 14:32 |
jeff |
flpl public machines offline, printing not working, staff computers can access the internet. |
| 14:32 |
jeff |
i'm guessing vpn burp, since pcc/etc haven't reported similar (so likely not pcr failure) |
| 14:32 |
jeff |
logging into meraki now |
| 14:33 |
jeff |
er, aerohive. |
| 14:33 |
* dbs |
feels for jeff |
| 14:33 |
jeff |
and... wrong window. |
| 14:33 |
jeff |
hi, #evergreen |
| 14:34 |
jeff |
:-) |
| 14:35 |
Dyrcona |
:) |
| 14:37 |
jeff |
and back up. |
| 14:37 |
jeff |
(since i know everyone here was waiting anxiously to hear) |
| 14:37 |
Dyrcona |
"And, there was much rejoicing!" |
| 14:40 |
|
bos20k joined #evergreen |
| 14:40 |
* tsbere |
ran a query to find all distinct Country names attached to user addresses in MVLC's database and dislikes that he got 24 results back for at most 5 countries |
| 14:45 |
tsbere |
And now I see that it is more like 35 if I don't clean up the various united states ones first <_< |
| 14:48 |
* Dyrcona |
just thought he had a 32MB file, but turns out it is only 3.2MB.... ls -lh is your friend. :) |
| 14:50 |
* dbs |
wonders how we get an opt_in_ws= (null) when a user gets opted in (thus angering the NOT NULL constraint on the actor.usr_opt_in table)... oh upgrades, you bring so many puzzles. |
| 14:55 |
tsbere |
dbs: Webstaff without a workstation, maybe? |
| 15:01 |
dbs |
tsbere: nope, just replicated it with my XUL staff client |
| 15:01 |
dbs |
And I haven't shared the webstaff url around with anyone here, trying to stabilize everything on day one of the upgrade first |
| 15:05 |
dbs |
there's code in auth.c and auth_internal.c for setting wsid, I hope our use of auth_proxy + LDAP isn't triggering a weird path where wsid doesn't get set. |
| 15:09 |
Dyrcona |
dbs: Did you just upgrade to 2.10? |
| 15:09 |
|
maryj joined #evergreen |
| 15:09 |
dbs |
Dyrcona: yep |
| 15:10 |
dbs |
oddly enough, just tried my master-from-about-a-month-ago test server and it propagated the wsid just fine |
| 15:10 |
Dyrcona |
Well, I would suspect the auth changes and something unforeseen with using LDAP. |
| 15:10 |
Dyrcona |
Any particular dot release? |
| 15:10 |
Dyrcona |
Like, 2.10.6? |
| 15:11 |
dbs |
2.10.6 indeed |
| 15:11 |
Dyrcona |
hmm.. Well, I'm not the expert in that area, but I would suspect something going on with the LDAP interaction, first. |
| 15:15 |
Dyrcona |
I'm not sure how much special code there is for LDAP, maybe none. |
| 15:15 |
dbs |
I would too, except we've been using auth_proxy and LDAP since 2.7. Maybe the rewrite for the bcrypt() auth missed that piece? But then I would have expected it to be backported from master (as it seems to work okay in master) |
| 15:17 |
dbs |
unless... /me tries logging in with his internal password, to let LDAP fall through |
| 15:17 |
Dyrcona |
Well, in that case, I'd diffs between master and 2.10.6 for the relevant bits, but I don't suspect that there are many. |
| 15:17 |
Dyrcona |
I'd [check] diffs... ;) |
| 15:18 |
dbs |
Yep, I think that's what's happening. When I registered the new workstation on the master system, I had to use the internal password to auth the first time around |
| 15:19 |
Dyrcona |
Ah, well, then that should be backported if it is missing. |
| 15:19 |
dbs |
When I used the internal password to auth on our 2.10.6 production system, I could opt-in the user. So I know to cast my baleful eye at the auth proxy and/or ldap code I guess |
| 15:19 |
dbs |
I'm surprised the lack of a wsid doesn't trigger more pain (like in all of the audit calls!) |
| 15:20 |
tsbere |
dbs: Due to the number of automated things that can generate audit info I believe they all have no issues with nulls |
| 15:23 |
dbs |
ah good |
| 15:23 |
dbs |
Maybe my workaround could be to drop the NOT NULL constraint on the opt_in_wsid column for now, until we get a proper fix... |
| 15:27 |
* dbs |
goes with a DEFAULT workstation named 'automatic for the people' instead |
| 15:27 |
Dyrcona |
dbs++ |
| 15:35 |
rhamby |
dbs++ : REM references should always be encouraged |
| 15:41 |
|
alynn26 joined #evergreen |
| 15:43 |
dbs |
heh, thanks :) |
| 16:05 |
|
bmills joined #evergreen |
| 17:35 |
|
bmills joined #evergreen |
| 20:51 |
|
bmills left #evergreen |
| 21:01 |
|
_adb left #evergreen |
| 22:26 |
|
StomproJ joined #evergreen |
