| Time |
Nick |
Message |
| 05:01 |
pinesol |
News from qatests: Failed Running perl live tests <http://testing.evergreen-ils.org/~live/test.24.html#2018-12-18T04:52:13,436767412-0500 -0> |
| 07:02 |
|
agoben joined #evergreen |
| 07:08 |
|
rjackson_isl joined #evergreen |
| 07:42 |
|
bdljohn joined #evergreen |
| 08:20 |
|
bos20k joined #evergreen |
| 08:20 |
|
Dyrcona joined #evergreen |
| 08:45 |
|
mmorgan joined #evergreen |
| 08:56 |
|
RMiller joined #evergreen |
| 08:59 |
RMiller |
Help! I'm stuck on OpenSRF installation on Xenial, getting this Perl error with the Jabber server: https://markmail.org/message/k26vrhvmfpzzkmwu |
| 09:02 |
|
remingtron joined #evergreen |
| 09:05 |
aabbee |
> which is supposed to be resolve dby making sure that the PATH=$PATH:/openils/bin ... |
| 09:05 |
aabbee |
what is the current value of $PATH? was .bashrc re-sourced, or did the user log out and back in, since modifying it? |
| 09:08 |
RMiller |
The last line of /opensrf/.bashrc is: export PATH=$PATH:/openils/bin |
| 09:08 |
aabbee |
what is the output of `echo "$PATH"`? |
| 09:10 |
aabbee |
actually, sorry, that's not the right question. if PATH hadn't been updated, then osrf_control would give a "command not found", but it isn't. |
| 09:11 |
RMiller |
The last bit of it says ...:/openils/bin anyway :) |
| 09:13 |
RMiller |
The ejabberd.yml file does have password authentication set to plain, with scram commented out, like it's supposed to... right? |
| 09:13 |
aabbee |
yes. |
| 09:15 |
aabbee |
the code says: |
| 09:15 |
aabbee |
$self->reader->connect($host, $username, $password, $resource); |
| 09:15 |
aabbee |
throw OpenSRF::EX::Jabber("Could not authenticate with Jabber server: $@") unless ( $self->reader->connected ); |
| 09:15 |
aabbee |
so $@ (perl var for error messages) is empty, but the reader is not connected. |
| 09:16 |
aabbee |
check through configuring the ejabber server once more, http://evergreen-ils.org/documentation/install/OpenSRF/README_3_0_2.html#_configure_the_ejabberd_server |
| 09:16 |
Dyrcona |
RMiller: Make sure that your passwords are correct in the opensrf_core.xml for each ejabberd domain and router. |
| 09:16 |
aabbee |
and note that there are instructions for different distros/versions are mixed together, so be careful to use the xenial ones |
| 09:16 |
bshum |
RMiller: Did you use any special characters in your ejabberd passwords? I think it only handles alphanumerics |
| 09:17 |
bshum |
So like if you have symbols, it might bomb unhappily |
| 09:17 |
aabbee |
and listen to Dyrcona and bshum, 'cause they're way smarter than me :-) |
| 09:18 |
Dyrcona |
Also, make sure ejabberd is running: pgrep -af beam |
| 09:20 |
RMiller |
The passwords are all alphanumeric. |
| 09:20 |
RMiller |
How many times should ejabberd show up in the output of pgrep -af beam? |
| 09:21 |
Dyrcona |
At least once. |
| 09:22 |
RMiller |
I don't exactly know how to read the output, but ejabberd is in there a couple times, so I assume it's running? |
| 09:23 |
aabbee |
there would be no output if it wasn't running. you can also check `systemctl status ejabberd`. |
| 09:24 |
Dyrcona |
Right. As long as it is there, it is running. |
| 09:24 |
RMiller |
The passwords are in there correctly, as far as I can tell; two are inside <password> tags and two are inside <passwd> tags. Is that right? |
| 09:27 |
aabbee |
yes. the <password> tags are for router users, <passwd> tags for opensrf users. note that these do not appear in the opensrf_core.xml file in the same order that the users were created in step 11. |
| 09:28 |
RMiller |
I gave them all the same password, so the order shouldn't matter... :) |
| 09:29 |
Dyrcona |
RMiller: For testing purposes, I use "password" so I don't have to change the configuration file. |
| 09:30 |
Dyrcona |
Did you restart ejabberd after making the changes to ejabberd.yml? |
| 09:31 |
Dyrcona |
You probably could not have registered the users unless you did, though. |
| 09:32 |
RMiller |
I'm sure I did (this has been a month-long on-and-off process, long story) but I just did again, and no change. |
| 09:33 |
Dyrcona |
OK. It can take a few tries to get the first installation working. After that, it gets easier. |
| 09:34 |
RMiller |
Yes, I remember from my first install three years ago how much "fun" this was... ;) |
| 09:35 |
Dyrcona |
check the ejabberd logs in /var/log/ejabberd |
| 09:36 |
|
yboston joined #evergreen |
| 09:42 |
pastebot |
"RMiller" at 64.57.241.14 pasted "ejabberd error log" (2 lines) at http://paste.evergreen-ils.org/14373 |
| 09:43 |
Dyrcona |
Well, that would indicate that the passwords are not right, most likely. |
| 09:44 |
Dyrcona |
At least the one for opensrf at private.localhost. |
| 09:45 |
Dyrcona |
I guess another thing to check is that the private.localhost and public.localhost names are properly defined in /etc/hosts. |
| 09:48 |
pastebot |
"RMiller" at 64.57.241.14 pasted "hosts" (4 lines) at http://paste.evergreen-ils.org/14374 |
| 09:49 |
RMiller |
The passwords are definitely right in opensrf_core.xml . Maybe I mistyped one in the configuration? How would I check/fix that? |
| 09:49 |
dbwells |
I've never put my finger on why, but it seems like every so often when setting up Evergreen I end up with a broken mnesia db for ejabberd. In those cases, the simplest "fix" I know is to simply stop all ejabberd/erlang processes, blow away everything inside /var/lib/ejabberd/ (leaving the dir), then restart ejabberd and re-register the users. There is almost certainly a better way, but the limited way we use ejabberd in Evergreen, it hasn't been |
| 09:49 |
dbwells |
worth me figuring it out yet :) |
| 09:50 |
Dyrcona |
RMiller: Maybe you registered the opensrf user with the wrong password for private.localhost. You can check your command line history: `history | grep ejabberdctl` |
| 09:50 |
Dyrcona |
dbwells: I've only ever had to do that if I've changed the hostname after ejabberd was installed. |
| 09:52 |
dbwells |
It could certainly be hostname related. |
| 09:54 |
RMiller |
...oops. My command line history does indeed indicate that I used the wrong password for private.localhost. |
| 09:54 |
aabbee |
ejabberdctl change_password opensrf private.localhost <password> # to change a pass |
| 09:54 |
Dyrcona |
aabbee++ |
| 09:56 |
RMiller |
...seg fault? |
| 09:58 |
aabbee |
ejabberdctl segfaulted? O_o |
| 09:58 |
Dyrcona |
Did you su ? |
| 09:59 |
Dyrcona |
I usually do sudo ejabberdctl .... |
| 10:00 |
RMiller |
Tried it as root, then as user with sudo. All gave me seg fault |
| 10:01 |
Dyrcona |
Try sudo -u ejabberd ejabberdctl ... |
| 10:01 |
RMiller |
Should I restart the service? |
| 10:01 |
aabbee |
it might be worth trying an `apt update; apt upgrade`, too |
| 10:03 |
RMiller |
Everything's upgraded |
| 10:07 |
Dyrcona |
RMiller: Anything in the ejabberd or syslogs about the segfault? |
| 10:09 |
RMiller |
Nothing in any of the ejabberd logs |
| 10:11 |
Dyrcona |
Not surprising: Segfaults often do not get logged anywhere, but worth a look. |
| 10:11 |
aabbee |
/var/log/syslog ? |
| 10:13 |
aabbee |
segfaults in ejabberdctl are weird, and indicate something is awry. but you might be able to make progress with the opensrf installation by updating the password in opensrf_core.xml (whatever bash history says it was initially set to) instead of fixing the password for the ejabberd user |
| 10:14 |
RMiller |
syslog has a lot of lines related to ejabberd that say apparmor="DENIED" Is that normal? |
| 10:14 |
Dyrcona |
Yes and no. |
| 10:15 |
aabbee |
what was the exact message from the segfault? https://ixnfo.com/en/troubleshooting-usr-sbin-ejabberdctl-line-428-14615-segmentation-fault.html (from 2017) claims to fix "/usr/sbin/ejabberdctl: line 428: 14615 Segmentation fault" by tweaking /etc/apparmor.d/usr.sbin.ejabberdctl |
| 10:15 |
Dyrcona |
It happened on older releases of apparmor on Ubuntu 16.04 when you used su to run ejabberd. |
| 10:15 |
Dyrcona |
I've never had it happen, but I run alway run ejabberdctl with sudo. |
| 10:16 |
Dyrcona |
bleh... I can English, I promise... :) |
| 10:16 |
RMiller |
It's line 428, yes. Let me give it a shot |
| 10:16 |
* aabbee |
was hoping an apt update would fix that |
| 10:17 |
RMiller |
Although I was able to register users initially, so that's kind of weird? |
| 10:17 |
Dyrcona |
RMiller: Did you try 'sudo -u ejabberd ejabberdctl change_password ...' |
| 10:17 |
Dyrcona |
aabbee: It should have fixed or not been a problem in the first place as I think that was fixed by the time Ubuntu 16.04.1 came out, but I could be wrong. |
| 10:18 |
RMiller |
Well, I thought I had... but I guess this time was the charm? It didn't give me any messages. |
| 10:19 |
aabbee |
whoo! \o/ (although Dyrcona's question remains unanswered) |
| 10:19 |
RMiller |
HEY! It's starting! |
| 10:19 |
Dyrcona |
Yay! |
| 10:20 |
RMiller |
(That was in reply to Dyrcona, actually :P) |
| 10:20 |
Dyrcona |
RMiler: You say everything is up to date, but did you restart the vm/computer after installing all of the updates? |
| 10:21 |
RMiller |
ooooh, so close. I tried to add 2 and 2 and the server timed out. |
| 10:21 |
RMiller |
Yes, I restarted after installing the updates. |
| 10:21 |
Dyrcona |
Are you running this on a watch? :) |
| 10:22 |
Dyrcona |
I've never had opensrf.math timeout, so not sure what's happening there. |
| 10:22 |
Dyrcona |
What does `osrf_control -l --diagnostic` say? |
| 10:23 |
Dyrcona |
How much RAM do you have? |
| 10:24 |
pastebot |
"RMiller" at 64.57.241.14 pasted "osrf_control diagnostic" (7 lines) at http://paste.evergreen-ils.org/14375 |
| 10:24 |
RMiller |
I lied. It didn't time out. It just didn't get a response. |
| 10:25 |
Dyrcona |
"router is not running" is bad. |
| 10:25 |
RMiller |
I'm an English teacher in the Caribbean, working with a Spanish-speaking IT guy, so I'm stretched right out to the very edges of my skills :D |
| 10:25 |
Dyrcona |
Nothing works without the router. |
| 10:25 |
RMiller |
I'm wondering if I registered all the users with the same wrong password |
| 10:25 |
Dyrcona |
Can you paste the output of `free -h`? |
| 10:26 |
Dyrcona |
Also, do you know if the operating system is 64-bit or 32-bit? |
| 10:26 |
pastebot |
"RMiller" at 64.57.241.14 pasted "freeeeeeee!" (3 lines) at http://paste.evergreen-ils.org/14376 |
| 10:28 |
Dyrcona |
How about `uname -p`? |
| 10:28 |
RMiller |
64-bit |
| 10:28 |
Dyrcona |
Ok. That's what uname -p would say. |
| 10:29 |
Dyrcona |
Yeah, I'd check all of the ejabberdctl register commands from the history to make sure about the passwords. |
| 10:31 |
RMiller |
I reset all the passwords to match the config file and now the router started :) |
| 10:32 |
RMiller |
Okay, restarting again got me math! |
| 10:32 |
RMiller |
4! Hooray! |
| 10:33 |
Dyrcona |
I'm concerned about the amount of RAM. I never go with less than 4GB. You may find yourself running out once you get PostgreSQL and everything else running. |
| 10:36 |
RMiller |
Ok. It's virtualized, so I'd have to ask the IT guy for that one. Are we talking increasing swap space or just actually adding another stick of RAM or what? |
| 10:37 |
pastebot |
"Dyrcona" at 64.57.241.14 pasted "Free on Ubuntu 18.04 test VM" (4 lines) at http://paste.evergreen-ils.org/14378 |
| 10:38 |
Dyrcona |
That's after having done very little, just looking up a book and editing it's information. |
| 10:38 |
Dyrcona |
Is this going to be for production use or just for testing? |
| 10:39 |
RMiller |
Production on a small scale- we're one K-12 school library. |
| 10:40 |
Dyrcona |
You'll probably want 8GB in the end. If it's virtualized, they should be able to increase the RAM rather easily, just a config change and VM restart. |
| 10:41 |
RMiller |
Ok. Thank you again for all your help! I really appreciate it |
| 10:42 |
Dyrcona |
You're welcome! |
| 10:43 |
|
RMiller left #evergreen |
| 10:44 |
|
khuckins_ joined #evergreen |
| 11:35 |
pinesol |
[evergreen|Jane Sandberg] LP1782652: Make template dropdown larger in Volume/Copy Editor - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=7b59293> |
| 11:41 |
|
yboston joined #evergreen |
| 12:03 |
|
jihpringle joined #evergreen |
| 12:19 |
|
mcgriff joined #evergreen |
| 13:22 |
|
khuckins joined #evergreen |
| 14:59 |
|
nfBurton joined #evergreen |
| 15:04 |
|
hbrennan joined #evergreen |
| 15:07 |
|
mmorgan joined #evergreen |
| 15:20 |
Bmagic |
Does Evergreen support Novelist "On the Shelf" ? Or has anyone got this working? |
| 15:21 |
Bmagic |
Instead of dropping javascript on the page, it looks like it needs to interact with an API and pull elements onto the OPAC but I know next to nothing about it |
| 15:24 |
Bmagic |
Clicking around, I might have found it. I guess it is just the same as Novelist select from a technical standpoint. Just drop the JS onto the page. |
| 15:25 |
|
mmorgan1 joined #evergreen |
| 15:47 |
|
khuckins_ joined #evergreen |
| 15:55 |
|
Dyrcona joined #evergreen |
| 16:15 |
Dyrcona |
Bmagic: That's correct, wrt Novelist On The Shelf. You also need to send them a more detailed data extract. |
| 16:15 |
Dyrcona |
My laptop's WiFi went out and I didn't notice that it had not reconnected to IRC. |
| 16:15 |
Dyrcona |
"Them" being EBSCO/Novelist. |
| 16:23 |
Bmagic |
Dyrcona: I see, thanks! Also - You're spying the channel without being logged in.... |
| 16:23 |
Bmagic |
Therefore: YOU'RE THE SPY |
| 16:24 |
hbrennan |
Bmagic: ha! |
| 16:24 |
Bmagic |
hbrennan: am I right or am I right? |
| 16:24 |
hbrennan |
Of course he's the spy |
| 16:24 |
Bmagic |
He's saying nothing. Just like a spy |
| 16:24 |
|
nfBurton joined #evergreen |
| 16:25 |
Bmagic |
Confirmed |
| 16:25 |
hbrennan |
Hey, so... In version 3.0.5 item barcodes aren't clickable. Wondering which version after fixes that? I can't seem to find anything even mentioning this bug |
| 16:27 |
hbrennan |
And by not clickable I mean when I click on the hyperlink barcode nothing happens and I get a "javascript;void(0)" when hovering over the barcode |
| 16:27 |
Dyrcona |
I'm always THE SPY, even when I'm not. |
| 16:28 |
Bmagic |
At least your are consistent |
| 16:28 |
mmorgan |
hbrennan: In which screen are you seeing that problem? |
| 16:28 |
Bmagic |
barcodes aren't clickable in the grids? All grids? |
| 16:28 |
hbrennan |
Anywhere |
| 16:28 |
hbrennan |
anywhere a Barcode column is |
| 16:28 |
hbrennan |
for items |
| 16:29 |
Bmagic |
When is your next upgrade cycle? |
| 16:29 |
hbrennan |
Trying to work on that right now |
| 16:29 |
hbrennan |
We got hectic so skipped the usual one in Nov |
| 16:30 |
hbrennan |
FYI I submitted my resignation yesterday so will be gone from the library March 1 |
| 16:30 |
hbrennan |
So I just want to get everyone on a stable release and train on the web client |
| 16:30 |
Bmagic |
3.0.9 has clickable barcodes (in the grid example I'm looking at) |
| 16:30 |
bshum |
autogen ? |
| 16:30 |
bshum |
In the past, after changes to org units, etc. and not running autogen, we weren't able to edit items, etc. |
| 16:30 |
Bmagic |
autogen! always seems to be the answer. The magic bullet almost every time |
| 16:30 |
bshum |
Autogen followed by apache restarts, etc. |
| 16:30 |
bshum |
Of course |
| 16:31 |
mmorgan |
Working in my 3.0.9 system as well |
| 16:31 |
hbrennan |
I can edit them, but I just have to copy the barcode into the item status screen manually |
| 16:31 |
hbrennan |
not idea |
| 16:31 |
hbrennan |
l |
| 16:31 |
hbrennan |
hmm |
| 16:31 |
bshum |
I just mean the links stop working |
| 16:31 |
hbrennan |
Yup |
| 16:31 |
bshum |
It sounds similar to what you're describing, just figured I'd mention it |
| 16:31 |
hbrennan |
bshum: So I need to do what then? |
| 16:32 |
hbrennan |
Sounds promising |
| 16:32 |
nfBurton |
Hey friends, can anyone point me to the definition for attrs.author. I don't see it in misc_utils |
| 16:32 |
hbrennan |
and explains why it wasn't causing an uproar |
| 16:33 |
bshum |
hbrennan: To run autogen.sh it's a script that you call up as the opensrf user on the server. And then restart apache, but doing a restart will drop people's current active connections. |
| 16:33 |
bshum |
So probably something you'd want to do a little afterhours (if there is such a thing in the world of today) |
| 16:33 |
hbrennan |
We (unfortunately) have to restart Apache often |
| 16:33 |
bshum |
Or just roll with it :) |
| 16:33 |
bshum |
People can refresh, it'll be fine... lolz |
| 16:33 |
hbrennan |
"I didn't do it!" |
| 16:34 |
hbrennan |
works often enough |
| 16:34 |
hbrennan |
:) |
| 16:34 |
hbrennan |
bshum++ |
| 16:34 |
bshum |
Only give me karma if I didn't just blow up your world by telling you to restart things :) |
| 16:35 |
hbrennan |
bshum: haha |
| 16:40 |
Dyrcona |
A reload might be all that's required. |
| 16:40 |
Dyrcona |
It's like a restart, except that the apache2 processes currently serving users will finish before quitting. |
| 16:42 |
bshum |
nfBurton: There's logic in misc_util for author. Above/around line 252 in master's version of that file |
| 16:42 |
|
bdljohn left #evergreen |
| 16:43 |
bshum |
@marc 100 |
| 16:43 |
pinesol |
bshum: A personal name used as a main entry in a bibliographic record. [a,b,c,d,e,f,g,j,k,l,n,p,q,t,u,4,6,8] |
| 16:43 |
bshum |
@marc 110 |
| 16:43 |
pinesol |
bshum: A corporate name used as a main entry in a bibliographic record. [a,b,c,d,e,f,g,k,l,n,p,t,u,4,6,8] |
| 16:43 |
bshum |
@marc 111 |
| 16:43 |
pinesol |
bshum: A meeting name used as a main entry in a bibliographic record. [a,c,d,e,f,g,j,k,l,n,p,q,t,u,4,6,8] |
| 16:43 |
nfBurton |
Thats for Graphic880s. |
| 16:44 |
bshum |
The graphic880s logic is used to link any of those against the author data for opac display. Like if you had an 880 entry in chinese it would display alongside the english text version |
| 16:44 |
bshum |
I think |
| 16:44 |
nfBurton |
So maybe I should explain. The year being in the query when clicking an author usually screws up the search so I have been removing the year from the query string |
| 16:44 |
nfBurton |
I cleared out on summary but still trying to figure it out for results |
| 16:45 |
nfBurton |
I just need to remove the birth/death year from the query |
| 16:45 |
nfBurton |
It's 100$d |
| 16:46 |
nfBurton |
The record summary was real easy to do since author is it's own file. But results just points to attrs.author and grabs everything |
| 16:51 |
bshum |
Well that does sound more difficult then |
| 16:52 |
nfBurton |
mmhmm |
| 16:52 |
nfBurton |
And the year often causes a blank search as it isn't always included |
| 16:52 |
bshum |
The function get_graphic_880s is defined at the top, and there are limits on which subfields get ignored, but I don't know if you'd want a "d" ignoring across the various other places the function gets used |
| 16:53 |
bshum |
Breaking it up away from how the authors get pulled in would be more complicated then |
| 16:53 |
nfBurton |
How does the attrs pull from the DB? |
| 16:55 |
nfBurton |
That worked adding d to the not(contains at the top. I thought that was for something else |
| 16:55 |
bshum |
It's probably something in EGCatLoader.pm perl side that works with the stuff in the template file |
| 16:55 |
nfBurton |
Oh Okay. Thanks |
| 16:55 |
nfBurton |
bshum++ |
| 16:56 |
bshum |
Putting that d in the ignore block will drop all subfield d from all the fields that call get_graphic_880s |
| 16:56 |
bshum |
Like title, authors, etc. |
| 16:56 |
nfBurton |
Oh, the title too..... |
| 16:56 |
bshum |
Publication year, etc. |
| 16:56 |
bshum |
So really you don't want to do that globally |
| 16:56 |
bshum |
Even though it works :) |
| 16:58 |
nfBurton |
Okay. Guess I could add a misc_util function, which I was hoping to avoid |
| 16:58 |
bshum |
It's kind of like you'd have to though. In order to preserve the original displayed content vs. the new variable you'd rather be searching on |
| 16:59 |
|
khuckins joined #evergreen |
| 17:01 |
nfBurton |
I thought one would exist already. Thanks for the sanity check. |
| 17:01 |
pinesol |
News from qatests: Failed Running perl live tests <http://testing.evergreen-ils.org/~live/test.24.html#2018-12-18T16:54:23,308313503-0500 -0> |
| 17:02 |
bshum |
It wouldn't be the first time we broke up an existing field definition into more granular pieces to achieve certain goals |
| 17:04 |
|
mmorgan left #evergreen |
| 17:54 |
|
aabbee left #evergreen |
| 18:27 |
|
sandbergja joined #evergreen |
| 18:48 |
|
sandbergja joined #evergreen |
| 23:18 |
|
sandbergja joined #evergreen |
