| # | Time | Nick | Message |
|---|
| # | 01:43:50 | atz has joined #evergreen |
| # | 07:54:29 | mck9 has joined #evergreen |
| # | 08:07:56 | sfortin has joined #evergreen |
| # | 08:14:26 | atheos has quit IRC |
| # | 08:22:00 | StephenGWills has joined #evergreen |
| # | 08:23:44 | StephenGWills | mornin ya tree-huggers! |
| # | 08:32:11 | grantjohnson has joined #evergreen |
| # | 08:32:14 | atheos has joined #evergreen |
| # | 09:00:18 | StephenGWills | when staff reports from the field that their client intermittedly produces screens, such as the advanced search, a MARC record editor add item sequence, then hangs on next call to adv. search screen with a blank screen. Are there troubleshooting tips documented anywhere I can fall back on. am kinda stump-ed. |
| # | 09:02:40 | atheos has quit IRC |
| # | 09:08:32 | atheos has joined #evergreen |
| # | 09:09:39 | jenny has joined #evergreen |
| # | 09:14:45 | Callender has quit IRC |
| # | 09:17:54 | gdunbar has joined #evergreen |
| # | 09:23:43 | Callender has joined #evergreen |
| # | 09:33:21 | eby_ has joined #evergreen |
| # | 09:41:39 | dbs has joined #evergreen |
| # | 10:02:32 | atz | we've had a lot of blank screen problems in trunk recently |
| # | 10:03:45 | atz | phasefx probably knows better than i do how to approach it... but you can try viewing the javascript console and watching for errors |
| # | 10:03:52 | phasefx | though I think those are with dojo |
| # | 10:04:21 | phasefx | white screens with the opac, or a xul interface, I'd look first to networking. check ping times |
| # | 10:07:19 | dbs | blank_screens-- |
| # | 10:07:59 | phasefx | well, I take that back with opac, I did get white screens on remote sites in the past with OPAC |
| # | 10:08:12 | phasefx | 0 byte documents, IIRC |
| # | 10:11:08 | StephenGWills | dbs: blank_screen =-1000 |
| # | 10:12:06 | phasefx | I'd try reproducing them and seeing if you can catch anything interesting in the apache access and error logs, and check for packet loss |
| # | 10:12:08 | StephenGWills | part of the problem is that i am not on-site so am talking a non-tech through js console... == helpdesk pain. |
| # | 10:12:22 | dbs | StephenGWills: that "hang" screen sounds familiar - you see part of the screen, but the language picker and some other elements don't get filled in, right? |
| # | 10:12:24 | StephenGWills | nod. will mess with it Jason, thanks. |
| # | 10:12:36 | dbs | StephenGWills: what version of dojo do you have installed? |
| # | 10:12:39 | StephenGWills | dbs: yes, that is the symptom |
| # | 10:13:11 | phasefx | also, any sort of transparent proxy can wreak havoc on things. Some think the staff client is a malicious web app and block the javascript :) |
| # | 10:13:15 | StephenGWills | am not strong Dojo, how to discover version? |
| # | 10:13:46 | dbs | check /openils/var/web/js/dojo/dojo/dojo.js.uncompressed.js and look for major / minor version numbers |
| # | 10:13:49 | atz | sometimes i think the staff client is a malicious web app too :) |
| # | 10:13:57 | StephenGWills | oh well if we can bolox a firewall we will, it's one of our specialties ;-) |
| # | 10:14:00 | phasefx grins, "I await your patch |
| # | 10:14:12 | dbs | StephenGWills: I was battling that problem back in the fall. at the time, i was convinced that it was a problem with apache requests dying because by default we were killing them off every 1000 requests |
| # | 10:14:42 | dbs | but changing that so that the apache processes lived for 100,000 requests didn't make a difference |
| # | 10:15:19 | phasefx | my latest pain/discovery with the staff client is that recent versions of Mozilla can't print to Generic/Text printers |
| # | 10:15:25 | StephenGWills | Dojo ver 1.3.2 null flag |
| # | 10:15:26 | dbs | but something in the last few months (whether a new dojo version, or an upgrade to 1.6.0.3+patches) seems to have resolved it |
| # | 10:15:40 | dbs | StephenGWills: yeah, okay, that's what we're running. hmm. |
| # | 10:17:45 | StephenGWills | I need to have staff track ping a little more systematically as well. I have the feeling that the network, especially their wireless is "iffy" but I am reluctant to let "network problems" sweep a real problem under the rug. |
| # | 10:18:52 | StephenGWills | thanks all, I'll gather more useful data |
| # | 10:19:00 | phasefx | NAT tables getting filled up can also be a problem if using cheap firewall/routers |
| # | 10:19:35 | StephenGWills | so, is the consensus that this is a ghost that kinda goes away and we like that? |
| # | 10:20:40 | phasefx wishes we could find root cause for every problem. Not enough info for this one |
| # | 10:23:00 | phasefx | anyone play with multiple versions of xulrunner on the same system and see one refuse to load the client after a while? |
| # | 10:24:53 | phasefx | weird, I recreate the build/ directory it's pointing to and it starts working again |
| # | 10:28:53 | phasefx | it's like once one version of xulrunner has touched the build/ directory, it calls dibs, and other xulrunners won't load it, even if no processes are running |
| # | 10:29:00 | BobW has joined #evergreen |
| # | 10:30:44 | phasefx | hrmm, compreg.dat and xpti.dat and extensions/ gets created. I think the middle one is compiled xul. Maybe some incompatabilities between these major versions of xulrunner |
| # | 10:32:28 | phasefx | compreg.dat is the culprit (thanks #evergreen for letting me rubber-duck) :) |
| # | 10:35:31 | phasefx | so we should have our installer delete those files when installing over an existing client installation |
| # | 10:36:18 | _bott_ has spent more hours than I care to count installing different xulrunners under OSX |
| # | 10:36:33 | emrikol | If anyone knows real quick... what method would I call to get info on the Internet:filtered/unfiltered/no access? |
| # | 10:37:07 | StephenGWills has so much to learn. not familiar with "rubber-ducking" tho ernie and bert tell me it's "so much fun"? |
| # | 10:37:29 | emrikol | http://en.wikipedia.org/wiki/Rubber_duck_debugging :D |
| # | 10:37:32 | phasefx | emrikol: it's .net_access_level on user objects, so whichever method retrieves a user |
| # | 10:38:05 | dbs | phasefx: always glad to serve as a sounding board |
| # | 10:38:19 | phasefx squeeks some sonar at dbs |
| # | 10:38:22 | dbs | _bott_: just install windows on your mac and you'll be fine |
| # | 10:38:36 | _bott_ goes screaming in horror at the thought |
| # | 10:38:47 | StephenGWills | rubber ducking++ although I will use my rubber penguin thanks ;) |
| # | 10:39:11 | phasefx | emrikol: that field will give an id referencing the config.net_access_level table |
| # | 10:39:16 | _bott_ | ...but already have it in Parallels and Vbox |
| # | 10:40:09 | phasefx | emrikol: open-ils.actor.net_access_level.retrieve.all to get those values with labels |
| # | 10:41:01 | dbs | phasefx: do you think we could add an ou_setting to trunk that would let us specify preferred skin for the staff client, rather than manually modifying constants.js on each individual workstation? |
| # | 10:41:30 | phasefx | emrikol: though if you're hacking the client, those are fetched during login and could be retrieved with JSAN.use('OpenILS.data'); var data = new OpenILS.data(); data.stash_retrieve(); data.hash.cnal and data.list.cnal |
| # | 10:42:15 | phasefx | dbs: sounds feasible. I know some folks are using apache URL rewriting as a workaround |
| # | 10:42:35 | dbs | phasefx: okay. another hackfest possibility! |
| # | 10:43:00 | dbs got trunk running again last night on a debian lenny image; only took a couple of hours (whee!) |
| # | 10:43:31 | phasefx | dbs: maybe having the entire menu system built from remotely retrieved data (even we can't just a remote xul file because of the opac wrapper) could be a hackfest item |
| # | 10:43:32 | emrikol | phasefx: No, luckily enough this has nothing to do with the client :) I'm wanting to integrate the "filtering" info in with our public computer scheduler |
| # | 10:44:58 | dbs suspects his hackfesting will be filing down the rough edges on the self-serve password reset |
| # | 10:50:45 | dbs | Or turning all of the current inline print()-based web pages (slimpac and the like) into TT2-based pages |
| # | 10:55:06 | berick | dbs: re:trunk, did you use the script or do it manually? |
| # | 10:55:18 | dbs | berick: just did a clean install |
| # | 10:55:46 | berick | dbs: heh.. re:trunk, did you use the script or do it manually? |
| # | 10:55:48 | dbs | (manually, if that's what you're asking) |
| # | 10:55:55 | berick | yeah |
| # | 11:00:49 | dbs | still on dojo 1.3.x for trunk, right? |
| # | 11:01:20 | berick | dbs: yeah |
| # | 11:01:53 | berick | though, i didn't have any noticible probs with 1.4 |
| # | 11:02:07 | berick | i believe i recall mike saying he saw some funkiness |
| # | 11:44:57 | miker_ | so ... just to reiterate ... IE -= 2^32 |
| # | 11:45:08 | miker_ | "XML spec? what's that" |
| # | 11:46:45 | miker_ | (in related news, I found why IE hates AutoIDL) |
| # | 11:46:51 | dbs | kbeswick: speaking of which -- if you're working on btresult, you might want to take a look at it in IE. It is... humourous. |
| # | 11:48:17 | StephenGWills | did anyone ever figure out the boot_Dynaloader issue? because i am trying to get ready for Grand Rapids, you KNOW it decided to visit me today. doh! |
| # | 11:49:42 | kbeswick | dbs: just was loading up winxp vbox as you typed that |
| # | 11:50:17 | kbeswick | ...which does not appear to want to connect to the internet. bah! |
| # | 11:59:06 | _bott_ | StephenGWillis: regarding plperl for postgres? |
| # | 12:04:03 | StephenGWills_ has joined #evergreen |
| # | 12:04:37 | _bott_ | StephenGWills_: was that boot_Dynaloader related to plperl for postgres? |
| # | 12:04:48 | StephenGWills_ | yes. |
| # | 12:04:55 | _bott_ | postgres built from source? |
| # | 12:05:02 | StephenGWills_ | i am also using the stow 8.2 postgresql |
| # | 12:05:05 | StephenGWills_ | yes |
| # | 12:05:31 | _bott_ | I received it when building 8.3.9 missing some configure options. |
| # | 12:05:59 | StephenGWills_ | ok... i have a lead i am going after will post if it solves it |
| # | 12:06:13 | _bott_ | I should say after building, it ran fine, but wouldn't take plperl and friends |
| # | 12:07:07 | _bott_ | ./configure --with-perl --with-openssl --with-libxslt --with-libxml --enable-integer-datetimes seemed to work well |
| # | 12:07:39 | dbs | StephenGWills: you're probably connecting to 8.3 instead of 8.2 or something like that |
| # | 12:08:19 | dbs | we've seen this problem recently where 8.4 was installed by default, and 8.3 was installed after, and the wrong port was getting connected |
| # | 12:08:34 | StephenGWills has quit IRC |
| # | 12:08:34 | StephenGWills_ is now known as StephenGWills |
| # | 12:24:40 | zed2 has joined #evergreen |
| # | 12:24:40 | StephenGWills | dbs: you may be right, I jsut manually restarted all systemmatically and problem went away. I tried tweaking shared_buffers the other day and probably restarted the wrong pg_ctl |
| # | 12:35:12 | StephenGWills_ has joined #evergreen |
| # | 12:37:37 | phasefx | dbs: it just so happens something I'm working (local customizations for sound effects) can be leveraged to change the OPAC skin defined in chrome |
| # | 12:37:57 | phasefx | my tactic is to xhr and eval /xul/server/skin/custom.js |
| # | 12:38:21 | StephenGWills has quit IRC |
| # | 12:38:21 | StephenGWills_ is now known as StephenGWills |
| # | 12:39:48 | phasefx | and do that with every remote interface, and with menu_frame.xul and opac.xul for chrome |
| # | 12:49:40 | phasefx | went ahead and committed that; now I get to see how git stash will work if the files stashed were modified a bit |
| # | 12:56:58 | zed2 has left #evergreen |
| # | 13:02:42 | jenny has left #evergreen |
| # | 13:29:21 | atz | phasefx: verdict? |
| # | 13:30:00 | phasefx | it did register a conflict where I couldn't really see a difference, edited the file and moved on. Didn't see any --continue type command or anything |
| # | 13:33:39 | dbs tried "svn merge" last night for the initial cut of the self-serve password stuff, then went back to "svn diff -r ####:HEAD" and hand-editing the diff file |
| # | 13:33:55 | dbs | and phasefx++ |
| # | 13:34:18 | dbs | phasefx: although - is that per-OU-ish? |
| # | 13:35:12 | phasefx | dbs: it _could_ be, nothing precluding you from pulling an org unit setting within custom.js and doing what you will with it |
| # | 13:35:49 | dbs | phasefx: just thinking that an OU setting would make things pretty simple, vs. having to edit another JS file |
| # | 13:37:44 | phasefx | for what I'm doing, a server-side javascript file is sufficient (they're wanting global changes). I think you're right with ou setting for opac skin. Just realized that this _could_ be another mechanism for you |
| # | 13:39:51 | StephenGWills has quit IRC |
| # | 13:42:54 | phasefx | for those 4 url constants, we could do something like if (data.hash.aous['opac_skin']) { url = url.replace(/default/,data.hash.aous['opac_skin']); } wherever they get used |
| # | 13:43:59 | phasefx | having the skin be a user option in My Account : Settings might also be interesting :) |
| # | 13:45:27 | BobW has quit IRC |
| # | 14:07:36 | brendan_bywater has joined #evergreen |
| # | 14:10:56 | StephenGWills has joined #evergreen |
| # | 14:17:34 | pmplett has joined #evergreen |
| # | 14:30:31 | atheos has quit IRC |
| # | 14:33:51 | natschil has joined #evergreen |
| # | 14:36:29 | atheos has joined #evergreen |
| # | 14:39:31 | natschil has quit IRC |
| # | 14:41:37 | GMLGeek has quit IRC |
| # | 14:44:46 | GMLGeek has joined #evergreen |
| # | 15:04:53 | pmplett is now known as pmpafk |
| # | 15:04:55 | dbs | Umm - for the Developers Update on Thursday morning - is somebody specific already responsible for that? I'd hate for a last-minute scramble to pull something together to get in the way of the pub crawl... |
| # | 15:06:35 | _bott_ | dbs: a little bird told me that miker_ and berick would be "in the know" |
| # | 15:06:42 | dbs imagines a comedy routine involving a terminal and a developer repeatedly running "svn update" |
| # | 15:07:00 | _bott_ | ...hopes she told them too |
| # | 15:09:09 | phasefx | don't we have a precedent for the developers not knowing anything about these developer updates? :) |
| # | 15:09:42 | _bott_ | can you really call once a precedent? |
| # | 15:11:27 | phasefx | I'd hope not in this case, but I think you can |
| # | 15:12:18 | gmcharlt | since there's no Supreme Court of Evergreen, I think we're safe and can ignore the precedent |
| # | 15:14:12 | dbs | we could brainstorm a list of "stuff what we should communicate" on the wiki: maybe with a list of links to show off features that might be visible on a dev server where that's appropriate |
| # | 15:14:20 | phasefx | plenty of judges, though |
| # | 15:14:37 | dbs | goal: force miker_ to spell words like "philosophy" on the fly :) |
| # | 15:15:00 | dbs hopes miker_'s cold is getting better |
| # | 15:15:18 | berick | dbs: yes, what you said, yes |
| # | 15:15:39 | miker_ | _bott_: was that little bird ... you? ;) |
| # | 15:15:53 | berick | miker_: so, what are we going to talk about? ;) |
| # | 15:16:10 | miker_ | berick: dunno ... let's ask the bird |
| # | 15:16:13 | berick | i there's a 50 minute window for the EG update |
| # | 15:16:19 | _bott_ | miker_: no, actually she's on the ESI payroll ;) |
| # | 15:16:20 | miker_ | gah |
| # | 15:16:30 | miker_ | _bott_: well, not for long! |
| # | 15:16:32 | miker_ | ;) |
| # | 15:17:04 | _bott_ | I believe that 50 minutes includes Bob's "the State of Evergreen" bit too |
| # | 15:21:24 | grantjohnson has left #evergreen |
| # | 15:21:48 | miker_ | berick: we could just read off the 1.6 feature list |
| # | 15:22:04 | berick asks bob how long his session will take |
| # | 15:22:29 | miker_ | "bob, can you talk for 45 mins?" |
| # | 15:24:44 | dbs | miker_: you mean, can he stop after 45 minutes? |
| # | 15:24:54 | berick chuckles |
| # | 15:24:55 | phasefx grins |
| # | 15:27:08 | dbs | 1.6 feature list would be reservations, self-serve password reset, and...? For acq stuff, you could do a super-minimal skim and direct people to berick's talk at 11:30 |
| # | 15:28:11 | sfortin has quit IRC |
| # | 15:31:08 | berick | dbs: right, and then a whole bunch of other features will be discussed at the kcls lunch-time thing |
| # | 15:31:24 | berick | for 2.0 |
| # | 15:50:02 | dbs | I could provide a slide or two (or a wiki page, or blog post) describing the state of the self-serve password reset feature. |
| # | 15:50:38 | _bott_ | dbs: "slide or two" just screams lightening talk |
| # | 15:55:49 | berick | _bott_: for the lightning talk, are you recommending usb drives for shuffling our "slide or two"'s? |
| # | 15:57:16 | _bott_ | berick: sure, or we can get them on the machine ahead of time if you've got it ready (e.g. Wed or Thur) |
| # | 15:57:25 | berick | k, thanks |
| # | 16:08:11 | jeff prepares to depart cleveland for grand rapids |
| # | 16:08:43 | dbs prepares to have a gut-wrenchingly early wake-up tomorrow to start the drive to grand rapids |
| # | 16:09:27 | gmcharlt | dbs: how early? |
| # | 16:10:16 | dbs | gmcharlt: oh, 5:30 or so. which is about 2 hours before normal. I'm a bit of a drama queen. |
| # | 16:11:39 | gmcharlt | 5 a.m. for me |
| # | 16:11:45 | gmcharlt takes dbs's crown |
| # | 16:12:18 | dbs | gmcharlt: do you have a 9-hour drive as well? |
| # | 16:13:13 | gmcharlt | dbs: six hours of flying |
| # | 16:13:40 | dbs | gmcharlt: man, that's no fun |
| # | 16:14:13 | berick | dbs: wow, i was wondering if you were going to drive. |
| # | 16:14:22 | berick | 9 hours.. ugh |
| # | 16:14:40 | gmcharlt | dbs: yeah, 9 hours of driving is no fun at all |
| # | 16:14:46 | berick | gmcharlt: where is your layover, atl? |
| # | 16:14:52 | gmcharlt | berick: yep |
| # | 16:15:13 | dbs | oh, it's going to be fun. kbeswick is with me from the start, and we're picking up r123 in Sault Ste. Marie at the 3.5 hour mark. ROAD TRIP |
| # | 16:15:14 | berick is in detroit, i think |
| # | 16:15:26 | gmcharlt | actually, IIRC, I'm on the same GTL to Grand Rapids run as many other Equinoxers |
| # | 16:15:34 | dbs | berick: avoid that cloud eh? |
| # | 16:16:15 | gmcharlt | dbs: just remember that doing hackfest and road trip simultaneously is dangerous, at least if you're the driver |
| # | 16:16:20 | berick | dbs: yeah, i'm fond of flying, especially with.. clouds |
| # | 16:17:07 | berick | too bad it's a 13-hour haul from here by car |
| # | 16:20:58 | berick | s/fond/not fond/ |
| # | 16:23:34 | dbs | okay, time to get home and pack. Safe travels, everyone |
| # | 16:24:06 | dbs | _bott_: oh - are the sessions going to be recorded? |
| # | 16:24:48 | _bott_ | dbs: yes, but don't expect hi-def well edited ;) |
| # | 16:25:15 | dbs | _bott_: awesome. then I don't need to offer to bring my video camera + tripod to try and tape some sessions for the Dutch |
| # | 16:25:37 | _bott_ | dbs: nope, we'll have something for them |
| # | 16:26:00 | _bott_ wonders if they wanna skype in the middle of the night |
| # | 16:26:23 | dbs thinks he recalls a song by that title |
| # | 16:33:49 | dbs has quit IRC |
| # | 16:52:20 | gdunbar has quit IRC |
| # | 17:24:54 | eby_ has quit IRC |
| # | 17:45:43 | brendan_bywater has quit IRC |
| # | 17:58:52 | pmpafk is now known as pmplett |
| # | 18:37:11 | brendan_bywater has joined #evergreen |
| # | 19:28:34 | jenny has joined #evergreen |
| # | 20:06:22 | bott-otr has joined #evergreen |
| # | 20:16:30 | jenny has left #evergreen |
| # | 20:27:47 | StephenGWills has left #evergreen |
| # | 22:36:30 | mck9 has left #evergreen |
April 10-13