| # | Time | Nick | Message |
|---|
| # | 00:33:52 | dbs | hmm. oils_utils.c seems to want limits.h included again |
| # | 00:41:28 | jeff | dbs++ # fedora/redhat porting |
| # | 00:43:23 | dbs | yay, clean compile and install! |
| # | 00:43:38 | dbs | leed is the dude behind the recent redhat stuff |
| # | 01:30:32 | dbs | whee, got the router, perl, and c services all working on Fedora. Time to go to sleep. |
| # | 01:30:36 | dbs has quit IRC |
| # | 02:40:24 | shadowspar has joined #evergreen |
| # | 04:01:41 | branflakes has quit IRC |
| # | 06:38:20 | arvind has joined #evergreen |
| # | 06:39:29 | arvind | i have encountered an error while autogen step in evergreen it says "Can't use an undefined value as an ARRAY reference at org_lasso_js.pl line 30" and stops |
| # | 06:41:48 | arvind has quit IRC |
| # | 07:17:55 | Granitize has joined #evergreen |
| # | 07:44:26 | sfortin has joined #evergreen |
| # | 07:44:57 | sfortin has quit IRC |
| # | 07:45:24 | sfortin has joined #evergreen |
| # | 07:47:11 | brendan_bywater has quit IRC |
| # | 07:47:26 | brendan_bywater has joined #evergreen |
| # | 07:48:54 | brendan_bywater has quit IRC |
| # | 07:55:59 | Granitize has left #evergreen |
| # | 07:56:22 | grantjohnson has joined #evergreen |
| # | 08:03:48 | atz has joined #evergreen |
| # | 08:13:15 | shadowspar has quit IRC |
| # | 08:47:18 | shadowspar has joined #evergreen |
| # | 08:47:56 | Melissa has joined #evergreen |
| # | 08:58:41 | alxp has joined #evergreen |
| # | 09:07:22 | kmlussier has joined #evergreen |
| # | 09:13:24 | mrpeters-isl has quit IRC |
| # | 09:14:00 | mrpeters-isl has joined #evergreen |
| # | 09:26:09 | dbs has joined #evergreen |
| # | 09:26:22 | dbs | miker_++ # for catching my stupid paste-o in 950 |
| # | 09:26:54 | dbs | /openils/bin/autogen.sh ran happily on Fedora 13, w00t |
| # | 09:29:09 | Dyrcona has joined #evergreen |
| # | 09:29:45 | Dyrcona | dbs: go get some rest. |
| # | 09:30:04 | dbs snores |
| # | 09:39:45 | kmlussier has quit IRC |
| # | 09:55:43 | bts has joined #evergreen |
| # | 10:08:00 | Dyrcona is finally having success loading bibs into trunk. |
| # | 10:17:10 | granitize has joined #evergreen |
| # | 10:37:57 | rsinger has joined #evergreen |
| # | 10:40:00 | youdonotexist has joined #evergreen |
| # | 10:47:47 | atz | Dyrcona, what made the difference? |
| # | 10:52:41 | wjr has joined #evergreen |
| # | 11:07:17 | Dyrcona | I followed miker_'s suggestions to load 10,000 at a time and dropped a few indexes. |
| # | 11:07:47 | dbs | any postgresql.conf tweaks for things like commit checkpoints, etc? |
| # | 11:07:57 | Dyrcona | So far, 380,000 bibs have loaded. |
| # | 11:08:29 | Dyrcona | I was on vacation for two weeks, and then back 1 week before doing this, so I don't know if those changes were made. |
| # | 11:08:57 | Dyrcona | There's another person here who usually messes with the server stuff for me. I'll have to check the config. |
| # | 11:10:32 | Dyrcona | Just going by the timestamps on the config files, I'd say, "No." |
| # | 11:10:47 | dbs | okay |
| # | 11:10:51 | Dyrcona | However, he might be running the daemon with an alternate config specified. |
| # | 11:11:18 | dbs | you could connect to the server via psql and show server settings if you really wanted to find out :) |
| # | 11:11:25 | TonyG has joined #evergreen |
| # | 11:11:45 | Dyrcona | Yeah. It's using the config file that I looked at earlier. |
| # | 11:11:47 | TonyG | Hello EG comm |
| # | 11:12:20 | jeff | greetings, fellow michigan-dweller. |
| # | 11:12:42 | TonyG | Thanks Jeff |
| # | 11:12:47 | Dyrcona | The suggestion was to set the checkpoints to 100 or 50, right? |
| # | 11:13:30 | TonyG | Hey, anyone know how to get EG to work with Ubuntu 10.04? |
| # | 11:14:04 | jeff | TonyG: have you run into a specific problem? |
| # | 11:14:12 | miker_ | Dyrcona: "enough" ;) ... if you're seeing "checkpoint recycling too frequently" or similar in the error log on a regular basis, double what's there. lather, rinse, repeat |
| # | 11:14:37 | TonyG | Stuck at the libdbi drivers for Postgre not being located |
| # | 11:17:19 | jeff | TonyG: is this when you're running "make -f Makefile.install ubuntu-karmic |
| # | 11:17:46 | jeff | ", or at another point? Is there a specific error message? If so, you can paste it to http://paste.lisp.org/new/evergreen and we can take a look. |
| # | 11:18:39 | Dyrcona | miker_: haven't seen the message about checkpoints since right at the beginning of this process. |
| # | 11:19:25 | Dyrcona | miker_: log says they are occurring 12 seconds apart. does that give any indication of what value to try next? |
| # | 11:19:50 | TonyG | no, on step V of server install, opensrf user loading /conf reports "pgsql driver not installed?" |
| # | 11:22:33 | TonyG | i found the 8.3 drivers, but I don't know if the install script is pointing to the drivers directory correctly |
| # | 11:23:53 | miker_ | Dyrcona: what's the current checkpoint segments setting? |
| # | 11:24:15 | Dyrcona | miker_: default of 3 |
| # | 11:24:51 | TonyG | jeff: maybe something to do with Path? |
| # | 11:24:55 | jeff | TonyG: oh, and this is Lucid, so there's no Makefile.install target in 1.6 |
| # | 11:25:30 | miker_ | Dyrcona: I'd start with 32 if you can spare 33-ish * 16MB of space |
| # | 11:25:32 | jeff | TonyG: are you trying 1.6 or trunk? It's very possible that 1.6 will not work without some additional tweaking. |
| # | 11:25:56 | Dyrcona | miker_: thanks. i'll try that for the next load. |
| # | 11:26:33 | rsinger has quit IRC |
| # | 11:26:38 | miker_ | Dyrcona: if you're simply testing, and really want to speed things up, you can just turn off fsync ;) |
| # | 11:27:54 | Dyrcona | miker_: I'm trying to get a rough guess on how long this will take under real conditions, so that I can compare different load methods and options: copy from vs. insert, 15 simultaneous loads vs. 7, etc. |
| # | 11:29:59 | miker_ | understood. though, still, if you trust your hardware enough you /can/ turn off fsync for the initial load and then just restart with it on |
| # | 11:31:28 | kbeswick_ has joined #evergreen |
| # | 11:32:04 | TonyG | jeff: I used 9.10 karmic on the make |
| # | 11:32:22 | TonyG | jeff: version 1.6.0.4 |
| # | 11:35:32 | Dyrcona | miker_: I think the log segments issue is when the automated load manager starts up and initially bombs the database with 15 files. After that, it keeps 15 files loading at all times, but it only adds a new file as others finish or fail. |
| # | 11:36:02 | TonyG | jeff: I have 0.8.3 libdbi and drivers, what is "trunk"? |
| # | 11:36:15 | Dyrcona | miker_: I am going to leave the settings as-is to see if we get the same message when only loading 7 at a time. |
| # | 11:37:07 | miker_ | Dyrcona: well, sure, but more segments means less IO storms for heap data (and longer checkpoint times) which is the goal -- smooth, long heap checkpoints |
| # | 11:37:16 | jeff | TonyG: "trunk" is just the bleeding edge version of evergreen. checking out evergreen from svn. both evergreen 1.6.0.4 and evergreen "trunk" use libdbi 0.8.3 |
| # | 11:37:40 | Dyrcona | miker_: Duly noted. |
| # | 11:37:45 | jeff | TonyG: i'm going to try something here. take several minutes, but i can get back to you. |
| # | 11:38:31 | kbeswick has quit IRC |
| # | 11:38:38 | TonyG | jeff: i have 2 dirs, libdbi-0.8.3 and libdbi-drivers-0.8.3 from the gz files |
| # | 11:42:35 | TonyG | jeff: i also found three possible files in /usr/local/lib/dbd, named libdbdpgsql.a, ".la and ".so |
| # | 11:42:58 | TonyG | jeff: correct location of files? |
| # | 11:43:27 | jeff | TonyG: did you have any issue with step IV a and b? |
| # | 11:45:33 | jeff | TonyG: can you paste the output of your run of configure to the pastebin at http://paste.lisp.org/new/evergreen ? |
| # | 11:46:12 | jeff | is the error right after "checking for libdbi pgsql driver (dynamic load)"? |
| # | 11:47:26 | bts | Question, when clicking the "done" button in a patron's record after finishing checkouts, is it always supposed to print or present you with the setting to print? |
| # | 11:47:28 | TonyG | jeff: how do i tell which version of Postgre I have loaded, step IV says 8.3 restart, i might have a newer version of Postgre? |
| # | 11:47:42 | bts | Or does that sound like some sort of bug and it should only print if we click something else for it |
| # | 11:48:14 | jeff | TonyG: on your database server, dpkg -l postgresql |
| # | 11:48:52 | jeff | bts: in our experience, "done" from the checkout interface will present the standard windows print confirmation dialog. |
| # | 11:49:18 | phasefx | unless you have Auto-Print checked, which suppresses the dialog and prints automatically |
| # | 11:49:40 | TonyG | jeff: postgre is 8.4.4 |
| # | 11:49:45 | bts | I see. What would be the best way to disassociate that printing from the button you think? |
| # | 11:49:48 | jeff | unless the auto-print checkbox is unchecking itself. ;-) |
| # | 11:51:17 | phasefx | is that happening, or just used to happen? |
| # | 11:51:50 | jeff | TonyG: i don't believe that 1.6 has formal support for postgresql 8.4. that may not be THE problem, but it may be or become A Problem for you. :) |
| # | 11:52:24 | jeff | phasefx: i believe we had reports in 1.4 of one of the auto-print boxes unchecking in an unpredictable fashion. I will make inquiries. |
| # | 11:53:16 | phasefx | bts: I'd like to see something like what the new billing interface in trunk does. There's a menu-button called Receipt Options, which has checkbox options Receipt Upon Payment and Printer Prompt, and a Number of Copies field |
| # | 11:53:36 | TonyG | jeff: config.log to pastebin? |
| # | 11:54:40 | phasefx | so, we could replace the Auto-Print checkbox with a Receipt Options button. A Receipt Upon Done checkbox would determine whether the Done button prints or not, and the Printer Prompt checkbox would be the opposite of the old Auto-Print checkbox |
| # | 11:54:57 | bts | phasefx: Aha, that would be most interesting. |
| # | 11:55:40 | phasefx volunteers bts for getting his xul/javascript hands wet :) |
| # | 11:56:06 | bts | phasefx: Heh, only done some minor tweaking, but that does sound like it would be interesting to try out. |
| # | 11:56:12 | jeff | TonyG: that and/or the output of configure that was echoed to the console. |
| # | 11:56:38 | bts | Out of curiosity, would doing something on that scale require repackaging a new staff client to handle the buttons? Or are those pulled off the server end? |
| # | 11:56:53 | jeff | no firm rule of thumb, but those are probably server-side. |
| # | 11:56:56 | phasefx | it's all server-side |
| # | 11:57:07 | TonyG | jeff: easy way to copy config so I can paste? certain app to use? |
| # | 11:58:22 | jeff | TonyG: i usually hilight the text in whatever terminal i'm using to connect to the server, copy, paste. you can cause the config.log file to be displayed on the terminal with "cat config.log" or even just the last 50 lines with "tail -50 config.log" |
| # | 11:58:48 | bts | Neato, sounds like a new project to tackle then. |
| # | 11:58:57 | bts | Thanks jeff and phasefx |
| # | 12:00:01 | mck9 | dbs: I will add #include to oils_utils.c, but it may take a while because I have some other stuff in the hopper. If you get to it first that's fine too. Thanks for the heads-up. |
| # | 12:00:11 | phasefx | bts: examples from billing, http://svn.open-ils.org/trac/ILS/browser/trunk/Open-ILS/xul/staff_client/server/patron/bill2.xul#L149 http://svn.open-ils.org/trac/ILS/browser/trunk/Open-ILS/xul/staff_client/server/patron/bill2.js#L770 |
| # | 12:04:24 | lisppaste | TonyG pasted "jeff: Config.log file" at http://paste.lisp.org/display/111809 |
| # | 12:05:44 | TonyG | jeff: will be back soon, have to go check on something |
| # | 12:32:16 | miker_ | dbs: so, if I'm reading the LoC docs correctly, $0 is, in fact, repeatable! |
| # | 12:32:26 | miker_ | *headdesk* |
| # | 12:52:39 | miker_ | grabbing 0315 |
| # | 12:55:11 | pmplett has joined #evergreen |
| # | 13:17:17 | rsinger has joined #evergreen |
| # | 13:18:30 | rsinger has quit IRC |
| # | 13:19:32 | mrpeters-isl has quit IRC |
| # | 13:48:28 | gmcharlt has quit IRC |
| # | 13:48:28 | gmcharlt has joined #evergreen |
| # | 13:55:44 | bts has quit IRC |
| # | 14:09:44 | Granitize1 has joined #evergreen |
| # | 14:10:59 | Granitize1 has joined #evergreen |
| # | 14:23:32 | dbs | mck9: heh, I got it in there last night :) |
| # | 14:24:09 | dbs | miker_: it wouldn't surprise me if it didn't say repeatable before... darned untrustworthy web |
| # | 14:28:50 | TonyG | jeff: what did you find out? |
| # | 14:29:11 | jeff | TonyG: were you able to capture the output of configure in addition to config.log? |
| # | 14:30:12 | TonyG | jeff: is there an additional switch to run with configure to get the output? |
| # | 14:30:55 | jeff | the output is displayed to your terminal and that's it by default |
| # | 14:31:17 | jeff | another thing to check, what does your /etc/ld.so.conf.d/eg.conf look like? |
| # | 14:31:33 | dbs | TonyG / jeff: PG 8.4 should work fine on 1.6 |
| # | 14:31:38 | jeff | dbs: thanks! |
| # | 14:32:13 | TonyG | dbs: thanks |
| # | 14:36:02 | TonyG | dbs: do you know if 0.8.3 pgsql driver will work with PG 8.4? |
| # | 14:37:08 | jeff | libdbi 0.8.3 version number is unrelated to the postgresql version number. |
| # | 14:37:25 | jeff | it's just a coincidence that libdbi 0.8.3 and postgresql 8.3 are similar |
| # | 14:38:24 | lisppaste | TonyG pasted "jeff: Config Output" at http://paste.lisp.org/display/111817 |
| # | 14:39:14 | TonyG | jeff: thanks, just checking, what are the chances? |
| # | 14:39:39 | jeff | TonyG: you've confirmed that the dbi files are present in /usr/local/lib/dbd, but I'm still unclear on if you were able to create /etc/ld.so.conf.d/eg.conf in step IV as well as run the ldconfig command and restart postgresql. Can you confirm those steps, and/or re-run them and try configure again? |
| # | 14:41:48 | TonyG | jeff: the eg.conf file hasboth /usr/local/lib and /usr/local/lib/dbd |
| # | 14:42:12 | TonyG | jeff: running other items asked, standby |
| # | 14:44:47 | dbs | I'm with jeff on the ldconfig bit. |
| # | 14:45:11 | dbs | However, it's possible that libdbi packages might have inadvertently been installed, which would potentially mess things up a bit |
| # | 14:45:37 | TonyG | jeff: restarting PG 8.3 does not work since I have 8.4; ldconfig and restarting 8.4 works |
| # | 14:45:56 | jeff | TonyG: okay, that's expected. now, can you try the configure command in step V again? |
| # | 14:46:30 | TonyG | jeff: as opensrf user or as root? |
| # | 14:46:43 | jeff | as opensrf user. |
| # | 14:46:55 | jeff | before that... if you've not started yet... |
| # | 14:47:07 | jeff | can you run: dpkg -l libdbi0 |
| # | 14:47:11 | jeff | (as any user) |
| # | 14:47:28 | jeff | ideally you should see: No packages found matching libdbi0. |
| # | 14:47:40 | dbs | (in interesting Fedora news, I found that I didn't have to build libdbi / libdbi-drivers from source, which means other platforms probably don't have to either, but we'll worry about that either) |
| # | 14:47:54 | dbs | s/either/later/ |
| # | 14:48:03 | dbs | that config.log file doesn't look complete |
| # | 14:48:53 | TonyG | jeff: libdbi0 is version 0.8.2-3 |
| # | 14:49:34 | dbs | ahh, that's the problem. "aptitude remove libdbi0" should help |
| # | 14:51:11 | TonyG | jeff: libdbi0 removed |
| # | 14:51:33 | dbs | configure again? |
| # | 14:51:37 | jeff | i'd suggest re-running ldconfig again as root "just to be safe", then try the configure command as the opensrf user as specified in step V |
| # | 14:52:34 | TonyG | jeff: you did it! thanks.. big high-5 to you! |
| # | 14:52:48 | jeff | dbs gets the credit on that one :-) |
| # | 14:52:54 | jeff | dbs++ |
| # | 14:53:24 | TonyG | dbs: kudos, triple kudos |
| # | 14:55:55 | dbs | jeff++ # tagteaming |
| # | 14:57:24 | TonyG | tagteam: just ran the "make", what is normal exit? |
| # | 14:57:50 | dbs | looks vaguely errorish I suppose. there's no "Congatulations!" at the end |
| # | 14:57:53 | dbs | paste it? |
| # | 15:00:34 | TonyG | TripleT(Tony'sTagTeam): No real errors, it seems. Let me try the rest of the install. |
| # | 15:01:50 | mrpeters-isl has joined #evergreen |
| # | 15:14:59 | pmplett is now known as pmpafk |
| # | 15:19:30 | TonyG | jeff: what is the "dbname" in perl command? do I need the <>? |
| # | 15:20:52 | jeff | you do not. the <> are just a convention indicating a non-optional parameter/argument. |
| # | 15:21:44 | TonyG | dbname is evergreen? |
| # | 15:21:48 | jeff | use the name that you used when running the "createdb" command -- in the install instructions, this is: evergreen |
| # | 15:21:51 | jeff | yes |
| # | 15:22:07 | TonyG | cool, thanks. |
| # | 15:23:38 | rsinger has joined #evergreen |
| # | 15:23:49 | rsinger has quit IRC |
| # | 15:25:22 | rsinger has joined #evergreen |
| # | 15:37:06 | rsinger has quit IRC |
| # | 15:45:26 | alxp has quit IRC |
| # | 15:56:22 | rsinger has joined #evergreen |
| # | 15:59:15 | kbeswick_ has quit IRC |
| # | 16:03:16 | sfortin has quit IRC |
| # | 16:05:52 | Melissa has left #evergreen |
| # | 16:08:31 | rsinger has quit IRC |
| # | 16:17:07 | TonyG | jeff: i think i am ready to test |
| # | 16:19:06 | jeff | TonyG: congrats! |
| # | 16:21:35 | TonyG | jeff: thanks, what is a good way to test srfsh? |
| # | 16:25:00 | TonyG | jeff: trying to test jabber connection next, correct? |
| # | 16:27:22 | TonyG | jeff: logged into as a new user and copied the .srfsh.xml file as root to new user /home |
| # | 16:27:48 | TonyG | jeff: change something in .srfsh.xml file? |
| # | 16:28:33 | atheos has joined #evergreen |
| # | 16:33:20 | TonyG | jeff: ok, logged in as opensrf and copied srfsh.xml file works |
| # | 16:34:44 | TonyG | jeff: but times out after 60 seconds with a "Name: osrfMethodException Status: No authentication seed found. open-ils.auth.authenticate.init must be called first Status: 404" error |
| # | 16:35:02 | atheos | can anyone tell me, by viewing the results of action.hold_request, how I can determine that a user has an available hold? I had assumed that a record with a capture_time and no fulfillment_time would indicate that a hold was available, but I've got an example where that's not the case. any info is greatly appreciated |
| # | 16:35:36 | TonyG | jeff: bad ssl cert? |
| # | 16:35:53 | jeff | atheos: are you doing a direct database query or are you making an api cstore request? |
| # | 16:36:01 | atheos | direct database query |
| # | 16:36:12 | atheos | raw |
| # | 16:37:54 | jeff | you need a capture time without a fulfillment or cancel_time, AND you need to either check for a transit with a dest_recv_time or you need to check the status of the copy as being "on hold shelf" |
| # | 16:38:38 | jeff | that may not be 100% but should get you on the right path. then ask yourself why you're not making an api call to get the information. ;-) |
| # | 16:38:46 | atheos | ok, that's the missing ingredient |
| # | 16:38:51 | jeff | on, depending on your needs a report. |
| # | 16:39:00 | atheos | lots of reasons why I'm not using the API, won't go into that :P |
| # | 16:39:06 | jeff | TonyG: that's in response to a "login" in srfsh? |
| # | 16:39:47 | atheos | jeff - if an item is in transit, it shouldn't be on the hold shelf, right? can I check for "on the hold shelf" only? |
| # | 16:40:02 | jeff | you should be able to, yes. |
| # | 16:41:13 | TonyG | jeff: yes, login admin open-ils |
| # | 16:41:58 | jeff | TonyG: is memcached running, and are the opensrf services running? did you start them with osrf_ctl.sh? |
| # | 16:44:46 | TonyG | jeff: memcached is running |
| # | 16:48:51 | TonyG | jeff: opensrf is running already and was started with shell |
| # | 16:52:44 | jeff | TonyG: try this in srfsh: request open-ils.auth open-ils.auth.authenticate.init "test" |
| # | 16:53:27 | jeff | should give you a line with Received Data: "hash here", then a summary and return you to the prompt. it should take less than a second. |
| # | 16:54:54 | TonyG | jeff: yes, received with hash |
| # | 16:55:45 | jeff | was the "Request time in seconds" less than a second? if not, how long did it take? |
| # | 16:55:50 | Dyrcona has left #evergreen |
| # | 16:57:03 | TonyG | jeff: Request Time in seconds: 0.134982 |
| # | 16:58:51 | jeff | okay. try the "login username password" again and see if it hangs for 60 seconds again. |
| # | 16:59:13 | jeff | if it doesn't work, you may want to take a look at the last several lines in /openils/var/log/osrfsys.log |
| # | 17:01:11 | r123 has quit IRC |
| # | 17:02:23 | atheos | jeff, thanks for the help - I'm getting the results I was looking for |
| # | 17:02:38 | jeff | atheos: good! |
| # | 17:03:14 | TonyG | jeff: same login timeout, let me check log |
| # | 17:03:32 | atheos | added this to my query, no more results for items in transit: ( SELECT status FROM asset.copy WHERE id=action.hold_request.current_copy ) = '8' |
| # | 17:05:26 | Granitize1 has quit IRC |
| # | 17:06:57 | lisppaste | TonyG pasted "jeff: open-ils login" at http://paste.lisp.org/display/111820 |
| # | 17:10:30 | jeff | TonyG: those lines seem to be cut off after a certain length -- and there were probably relevant errors before that first line also. |
| # | 17:10:56 | jeff | TonyG: try "tail -f /openils/var/log/osrfsys.log" in one terminal while you try another login attempt. |
| # | 17:11:15 | jeff | i've got to run for a few, but will check back later. |
| # | 17:11:49 | TonyG | jeff: got to get home to the family, will try again tomorrow, thanks for all your help |
| # | 17:13:01 | TonyG | jeff: not bad for my first time, Linux, Ubuntu, Postgre, Evergreen and trying to find help (you), fun stuff, glad you are there |
| # | 17:13:50 | granitize has quit IRC |
| # | 17:14:06 | TonyG | now how to you logoff properly from freenode? anyone? |
| # | 17:14:29 | TonyG | quit: |
| # | 17:14:30 | moodaepo | TonyG: I just close the window |
| # | 17:14:45 | moodaepo | or / with out the or : )quit |
| # | 17:14:52 | moodaepo | oops or /quit |
| # | 17:14:59 | TonyG has quit IRC |
| # | 17:23:00 | pmpafk is now known as pmplett |
| # | 18:19:48 | dbs has quit IRC |
| # | 18:43:11 | youdonotexist has quit IRC |
| # | 20:24:05 | rsinger has joined #evergreen |
| # | 20:30:38 | dbwells_ has joined #evergreen |
| # | 20:34:03 | bradl_ has joined #evergreen |
| # | 20:34:57 | dbwells has quit IRC |
| # | 20:34:57 | bradl has quit IRC |
| # | 20:35:04 | dbwells_ is now known as dbwells |
| # | 20:50:34 | atz has quit IRC |
| # | 20:51:49 | atz has joined #evergreen |
| # | 21:06:52 | dbs has joined #evergreen |
| # | 21:08:34 | dbs | Neat. Running trunk opensrf, trunk evergreen, and dojo 1.4.3 on fedora 13, and loading the main opac page results in an autorefresh frenzy in both chrome and firefox |
| # | 21:08:41 | dbs | otherwise, everything is looking good :) |
| # | 21:09:16 | gmcharlt | well, we wouldn't want your computer to get out of shape for lack of exercise :) |
| # | 21:19:01 | bradl_ is now known as bradl |
| # | 21:21:22 | dbs | bradl! |
| # | 21:21:37 | dbs | chromium spins its wheels lots faster than firefox does :) |
| # | 21:24:31 | dbs | Oh fun. |
| # | 21:25:02 | dbs | If I disable the opacLocaleSelector, it looks like that fixes the problem. Maybe a dojo 1.3 -> 1.4 api change |
| # | 21:33:03 | dbs | Yep. Dropped in Dojo 1.3.3, and all works well. |
| # | 21:41:00 | dbs | I guess we need to enable xmlent for idlchunk |
| # | 21:46:37 | dbs | weird. the entities aren't getting resolved, even though they're in fm_IDL.dtd |
| # | 21:56:10 | dbs | Heh |
| # | 21:56:26 | gmcharlt | ? |
| # | 21:56:45 | dbs | When I cut chunks out of fm_IDL.xml, it returns 16592 bytes before slamming a <!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN"> |
| # | 21:56:54 | pmplett has quit IRC |
| # | 21:56:58 | dbs | if I cut out some more chunks, it returns the same number of bytes |
| # | 21:58:41 | dbs | methinks we have either a leak, or a limit on the number of entities that can be resolved |
| # | 21:59:30 | dbs | seems to be related to entity resolution, because I can drop the non-entityized version into place and it works fine |
| # | 22:12:11 | dbs | @later tell Dyrcona importing 1000 records via vandelay on the completely untuned postgresql 8.4 that comes with Fedora 13 took about 2 minutes |
| # | 22:12:11 | pinesol | dbs: The operation succeeded. |
| # | 22:14:06 | dbs | @later tell Dyrcona and yep, that's current trunk |
| # | 22:14:06 | pinesol | dbs: The operation succeeded. |
| # | 22:30:06 | lisppaste | dbs pasted "searching in trunk - unhappiness in open-ils.storage.biblio.multiclass.staged.search_fts.atomic" at http://paste.lisp.org/display/111826 |
| # | 22:32:12 | miker_ | dbs: 2min for full import, or just into a vand queue, or just from a vand queue into the catalog? |
| # | 22:32:21 | lisppaste | dbs annotated #111826 "Ah, I guess that's the problem: Class::DBI" at http://paste.lisp.org/display/111826#1 |
| # | 22:32:30 | dbs | full import |
| # | 22:33:32 | miker_ | dbs: re cdbi, complain to tony about the api changes in 3.0.x after 3.0.1 |
| # | 22:33:56 | miker_ | (such as, ya know, the search() method api ... ;) ) |
| # | 22:34:15 | miker_ | but I'm glad to see my version test works as intended! ;) |
| # | 22:34:42 | miker_ | (we could, of course, switch to Class::DBI::301frozen or whatever it's called) |
| # | 22:35:12 | dbs | I just hate having to install more prereqs from source. |
| # | 22:35:28 | dbs | 301frozen gets bug fixes? |
| # | 22:35:59 | dbs | http://search.cpan.org/dist/Class-DBI-Frozen-301/ ** UNAUTHORIZED RELEASE ** |
| # | 22:36:01 | dbs | I like that |
| # | 22:37:23 | miker_ | were you around for the "fun" of the split? |
| # | 22:37:52 | miker_ | read the POD ... the original author is ... unnice |
| # | 22:39:01 | dbs | I vaguely recall discussing this with you a few years ago, but never paid attention to the primary source |
| # | 22:42:30 | dbs | Does this have any impact on the C:DBI: plugin modules? |
| # | 22:43:38 | dbs | on fedora, at least, perl-Class-DBI-Pg also gets removed. bah. |
| # | 22:43:38 | rsinger has quit IRC |
| # | 22:44:51 | rsinger has joined #evergreen |
| # | 22:49:37 | miker_ | not sure on cdbi plugins |
| # | 22:49:56 | dbs | ARGH, kill me. Stupid dumbass test tries to connect to local Pg database and fails, wants envvars set. Let's crank that EG install failure rate up to 11! |
| # | 22:50:32 | dbs | ah, no, that's not the problem |
| # | 22:50:51 | dbs | t/11-triggers.t ....... mutiple trigger registration in one add_trigger() call is deprecated. |
| # | 22:50:53 | dbs | wheee |
| # | 22:51:11 | miker_ | re idlchunk, it's a specialization of xmlent, just not outputting classes that aren't named (or all, if none are named) IIRC, so shouldn't need both idlchunk and xmlent |
| # | 22:52:56 | dbs | right, I worked that out |
| # | 22:53:19 | dbs | but it seems to be choking on the complete entityized fm_IDL.xml |
| # | 22:56:58 | dbs | apparently the multiple trigger problem is a conflict with modern Perl versions |
| # | 22:58:25 | dbs weeps, force installing C:DBI:Frozen:301 is fine, but Class::DBI::Pg doesn't recognize that as C:DBI so it wants to install C:DBI |
| # | 23:01:51 | jeff | hello |
| # | 23:04:18 | miker_ | dbs: do we need class::dbi::pg? |
| # | 23:04:35 | miker_ | I don't recall that, but it's been forever |
| # | 23:04:51 | dbs | miker_: hmm, I had assumed that, but that could be a bad assumption |
| # | 23:05:23 | dbs | I was basing the fedora work on the debian prereqs, which include C:DBI:Pg |
| # | 23:05:55 | dbs will try without |
| # | 23:06:33 | rsinger has quit IRC |
| # | 23:08:06 | dbs | ack doesn't show any uses of it, that's promising |
| # | 23:22:57 | dbs | Yeah. Class::DBI::Frozen::301 works fine, as long as we change the "use Class::DBI;" calls to "use Class::DBI::Frozen:301;" (but _not_ the package or use base calls, C:DBI:Frozen:301 does its own hacking of its module names) |
| # | 23:23:30 | dbs | ahh, sweet search interface... |
| # | 23:31:50 | dbs | I guess we could use UNIVERSAL::Require to 'Class::DBI'->use() or 'Class::DBI::Frozen::301'->use() or die |
| # | 23:45:58 | dbs | yay, works |
| # | 23:49:21 | miker_ | whee |
| # | 23:49:24 | miker_ | dbs++ |
| # | 23:55:03 | dbs | miker_: your sharp eyes would be appreciated on r16811 |
| # | 23:55:12 | dbs | my dull eyes are going to go to sleep |
| # | 23:55:14 | miker_ | dbs: expect an authority related present in the morning |
| # | 23:55:25 | miker_ | 'nite. I'll skim it |
| # | 23:55:27 | dbs | Oh hot damn! I need to get that email out :) |
| # | 23:55:40 | dbs | thanks miker_ - enjoy the ALA show (assuming you're there) |
| # | 23:57:11 | miker_ | nope |
| # | 23:57:18 | miker_ | I'm holding down the fort |
| # | 23:58:09 | dbs | Ahh - do you get to go to Access then? Or mayhaps OLA next year? |
| # | 23:58:12 | miker_ | commit looks good to me, assuming ->use() returns false, which I guess it should (last lval in the .pm, or undef on fail, I would assume) |
| # | 23:58:28 | miker_ | I'm going to try for access |
| # | 23:58:42 | miker_ | and if ola doesn't overlap with c4l ... |
| # | 23:59:48 | miker_ | go sleep!I am too |
April 10-13