IRC log for #koha, 2009-06-23

All times shown according to UTC.

Time S Nick Message
12:05 ebegin good morning!
12:08 |Lupin| hi ebegin
12:14 Amit hi nicomo
12:14 nicomo hi (again)
12:14 nicomo sorry I have network issues : coming in and out
12:14 Amit hi Lupin
12:25 |Lupin| Hi |Lupin|
12:47 |Lupin| oops
12:47 |Lupin| hi Amit
12:51 |Lupin| Recently I asked here how one should proceed to represent a e-book that is present in several formats, givne that each format could comprise several files. I was told that one record for each format would be good policy, but I'm wondering how the records would be linked to each other ?
12:51 |Lupin| For one given book, you don't want to re-enter all the bibliographic information for each format, and you don't want to correct, say, a typo, in each format...
15:48 danny brendan, are you around?
15:49 brendan hey danny
15:50 danny hey brendan, I saw your tweets and wanted to take a look at those links to the page but it seems to not load for me, is it down at the moment?
15:50 brendan could be I will take a look.
15:51 brendan hey danny go ahead and try again.
15:52 danny looks like the same thing, I get a Network timeout
15:52 danny i got this a couple of weeks ago when I tried to take a look at the site too
15:52 brendan darn -- it loads up for me -- I will keep looking into this
15:52 danny hadn't tried again until today and thought it was weird that it was still happening
15:52 jdavidb works fine for me, too.
15:53 brendan yeah - i just tried another computer and it loaded up for me.
15:54 danny i've tried several computers, and none are able to load the pag
15:54 danny maybe a router between the two?
15:56 brendan could you try a traceroute and see where it dies?
15:58 brendan hey there jdavidb
15:58 jdavidb Howdy, brendan.  :)
15:59 danny
16:00 brendan danny can you get to the main page ok -- just not the blog ?
16:00 danny no main page doesn't load either
16:01 danny It isn't an address not found, but always get a Network Timeout after about a minute
16:02 brendan does the name resolve to an IP ?
16:02 brendan something like is what I'm getting
16:02 danny yep, it shows that in the trace route too
16:02 brendan ok -- I gotta run I will be back in about 20 to 30 minutes
16:02 danny ok
16:05 slef hi all
16:06 slef is there a reason why koha doesn't blank the local use 9 fields from staged imports?
16:06 gmcharlt slef: 9xx tags?  any, or just the ones mapped to items?
16:09 slef everything except 942 really
16:09 slef and 098
16:09 gmcharlt well, it's not correct behavior to always drop 9xx tags - library may want some/all of them
16:09 gmcharlt course, one could add an option to remove such tags
16:10 slef when would a library want some/all of them?
16:10 slef but a "remove from import" option may be a useful enhancement then?
16:11 gmcharlt slef: really depends on what the 9xx from the source are for; can't just unilaterally remove them
16:11 gmcharlt thus, a remove selected tags from import option would be a useful enhancement
16:14 slef 9xxs are meant to be local use only, aren't they? So very few libraries will be relying on values in there
16:17 gmcharlt slef: but it could be *their* local use - e.g., order information from MARC records sent by book vendor, information used by a local union catalog, bibliographic utility information, etc.
16:20 slef gmcharlt: I see. Suspect the default should be strip, shouldn't it?
16:22 gmcharlt slef: I'd suggest letting user set up a profile defining which tags they want removed
18:49 |Lupin| hi again everybody !
18:55 SirStan hi lupin
19:03 |Lupin| hi SirStan
19:26 |Lupin| bye all
19:32 chris morning
20:00 brendan morning chris
20:00 brendan :)
20:02 chris heya brendan
20:06 owen Hi chris, how's your new little bundle?
20:06 chris doing good, noisy sleeper, but good
20:06 chris and yours ?
20:07 owen Very good. Noisy sleeper? Meaning he sleeps, but noisily?
20:08 chris yeah lots of grunts
20:09 hdl_laptop hi chris.
20:09 wizzyrea mine used to snuffle quite loudly
20:09 hdl_laptop owen: my baby has been quite noisy sleeper too.
20:10 hdl_laptop chris: I have a question about franch translation.
20:10 chris righto
20:10 wizzyrea omg, so many koha babies :D
20:11 hdl_laptop mmm maybe kohaman from katipo could be coool in that purpose ;)
20:11 owen I happily consider all three of my babies Koha babies. My first was born just before we first switched to Koha.
20:13 hdl_laptop chris: there have been updates on french translation after the release and it seems there was some translated strings which passed untranslated
20:13 gmcharlt wizzyrea: it looks like cafepress does onesies, so you could shoot Tina Burger an email and have her add that as an option for the Koha ILS store
20:15 hdl_laptop chris: is pootle site uptodate with po files from official branch ?
20:15 chris should be be
20:16 chris generally though i go from pootle to git
20:16 chris not the other way
20:18 hdl_laptop chris: only two days before you updated french translation, i pushed a patch with french translations.
20:18 hdl_laptop And they were ok at that time.
20:19 chris it may be then that they were overwritten from pootle
20:19 chris ie pootle -> git
20:19 chris very very rarely to i go git -> pootle
20:25 hdl_laptop french update from Chris Cormack  Fri May 29 711aacd7b8018b4592ee39a2407651c236572e07
20:25 wizzyrea gmcharlt hehe I will check
20:26 chris yep, that will be from pootle
20:26 hdl_laptop french update from me : 1b7c8f84e1faf87b44ab4e224913cbfa5ef15b2c  Mon May 25 16:43:21 2009
20:27 hdl_laptop and yours have overwritten mine
20:27 hdl_laptop So I will try and revert french po files.
20:28 hdl_laptop And merge on pootle
20:28 chris yep, what will have happened is git will have seen that the ones in pootle were different
20:28 chris is it opac, or staff?
20:28 hdl_laptop both
20:28 chris if staff, send me the file and ill merge on the commandline
20:28 chris it is sllllooooowww on the web interface and locks pootle up
20:28 chris so ill do it commandline and add it
20:28 hdl_laptop ok sending you both.
20:28 chris cool
20:31 gmcharlt SirStan: say what about it?
20:35 richard hi
21:56 SirStan gmcharlt: some hint of what the difference is.. or atleast a one line description.
21:58 hdl_laptop Sirstan : the difference is all about cataloguing habits
21:59 hdl_laptop some ppl are used to UNIMARC way to input data (French and European)
21:59 hdl_laptop Others are more used to MARC21 (US and many other countries).
22:00 hdl_laptop It comes along with different cataloguing frameworks.
22:00 hdl_laptop good night
22:01 SirStan THanks hdl -- even that microsnippet would rock.
22:02 SirStan or do most people know what they are
22:10 chris not most
22:10 chris but lots
22:10 chris trained librarians would know
22:11 chris but there are lots of other people who work in libraries that wouldnt
22:12 chris ill add the files now hdl_laptop
23:00 pianohacker Does anyone know who the original author of the reports/ code is?
23:00 gmcharlt pianohacker: git blame help?
23:01 pianohacker Seems to point to paul and hdl, but I know the history was truncated by some pre-3.0 merges by jmf and paul
23:02 gmcharlt to dig further, check out the old Koha CVS repo on savannah
23:02 pianohacker Ahh, good idea
23:05 chris which reports code? the guided reports?
23:05 chris or the wizards?
23:05 pianohacker Nope, the canned reports
23:05 pianohacker Ahh, it's hdl
23:05 chris ahh cant blame me for those
23:05 chris hehe
23:05 chris i take responsibility for any mess you find in guided reports though
23:06 pianohacker Given the horrifying hacks I'm adding to the reports wizards to add some functionality, I'm not the one to say so
23:07 chris :-)
23:19 chris @later tell hdl_laptop i have updated pootle with the french .po files
23:23 chris i wonder if you can queue up messages
23:24 chris @later tell and pushed them up to git
23:24 chris gah
23:24 chris @later tell hdl_laptop and pushed them up to git
23:25 pianohacker Apparently only works if you've registered. Oh well
23:32 chris heh
00:01 pianohacker1 evening, all
03:11 Amit hi brendan, chris, mason
03:12 Amit good morning #koha
03:12 brendan heya amit
03:12 Amit hi richard, Jo
03:12 brendan I hope all is well
03:12 Amit yes
03:12 Amit any new from your side brendan
03:12 brendan not at the moment
03:12 brendan anything new with you?
03:16 Amit something adding MARC to text
05:22 chris evening
05:26 richard hiya chris
05:29 richard :0
05:29 richard ack
05:29 richard :)
05:29 mason there we go ;)
05:43 chris heh
05:48 chris @wunder wellington, new zealand
05:49 chris duh, munin isnt in the channel
05:53 richard feels like there'll be a frost tonight
05:54 chris yeah clearish skies
06:43 chris back
06:49 magnusenger Good morning, all!
06:50 magnusenger Mr. Poulain is writing about Koha and CAS on the mail list - I feel like I should know what CAS is, but I'm coming up blank... Anyone care to enlighten me?
06:51 nicomo magnusenger: CAS is an sso solution
06:51 chris God morgen
06:51 chris or is it Godmorn
06:51 magnusenger God morgen, chris!
06:51 chris or maybe Go'mårra
06:51 chris :)
06:51 nicomo magnusenger see
06:51 magnusenger :-)
06:52 nicomo wow chris i'm much impressed by your norwegian
06:52 magnusenger chris: depends on how formal and distinct you want to be...
06:53 magnusenger nicomo: you should take note and pick up some things that might come in useful this fall... ;-)
06:53 chris i cheated nicomo
06:53 magnusenger Thanks for the enlightenment, nicomo!
06:53 chris[…]rs/gmorning.htm#N
06:54 nicomo chris: I wouldn't advice you follow this web site for the French at least
06:54 chris yeah, or the maori
06:54 magnusenger Wow, including the correct form for Sortland is kind of hilarious
06:55 magnusenger Sortland is a tiny place even further north than me...
06:55 Amit hi magnusenger
06:55 magnusenger hi Amit
06:57 chris i wonder how you say it in sami?
06:57 magnusenger re CAS: sounds nice! but here in norway we, of course, have go and invent our own way of doing these things...
06:57 nicomo magnusenger: I thought further north from you people just fell off the earth
06:57 chris Buorre idit
06:58 magnusenger nicomo: almost, but i'm still more than 600km from the Nort Cape
06:59 chris when i worked for the maori studies department, we had some sami people visit
06:59 magnusenger chris: i can imagine the sami and the maori have some things to talk about, yes
06:59 chris to take a look at the kohanga reo, and kura kaupapa maori (total immerision maori schools)
07:02 chris[…]on-library-geeks/
07:03 chris nice blog today
07:06 chris speaking of CAS, someone was talking about adding openid to the opac, allowing users to login using their openid, rather than a new username and password
07:06 chris i think that would be pretty cool, certainly for universities
07:12 magnusenger openid is a good thing, but then it would have to be tied to their user account somehow?
07:12 chris yep
07:14 magnusenger how would they do that?
07:14 chris the way i was thinking was that you could use your library cardnumber, to request an email be sent to you, which contained a link with and md5 hash, that you could go to, to link your openid to your account
07:14 magnusenger sounds good!
07:21 magnusenger ...and rather obvious... the midnight sun kept me up all night, so my head is not fully working today, methinks...
07:21 chris hehe
07:38 magnusenger anfscd: I did a successfull import of about 400K records into Koha 3 w/Zebra. Then I had to re-import because of some errors in the first batch. Then I ran into some problems with dropped connections etc. Now is taking forever "deleting biblios", and i'm getting this: "DBD::mysql::db do failed: The total number of locks exceeds the lock table size at /usr/share/koha/bin/migrati​on_tools/ line 100." i
07:51 chris back
07:55 |Lupin| hello all
07:55 |Lupin| hi chris
07:58 chris magnusenger: that doesnt sound goo
07:58 chris d
07:59 magnusenger i was afraid someone would say that... ;-)
08:00 chris lemme just look what line 100
08:01 chris what version of Koha is it magnusenger ?
08:01 magnusenger chris:
08:01 chris ta
08:03 chris can you do mysqladmin -uroot -p show processlist
08:03 chris i suspect it is the truncate commands locking things up
08:03 magnusenger also, at one point i thought the import was good, but then the rebuild_zebra reported about 900K records...
08:03 magnusenger chris: will do
08:07 magnusenger I get this:
08:09 chris right so its all finished now ... ok
08:09 chris so heres what we can do
08:09 magnusenger I got the same error again, but this time on line 101
08:09 magnusenger and then it says "10 MARC records done in 0.839056968688965 seconds"
08:09 magnusenger but it actually took about two hours
08:09 magnusenger (i tried importing 10 records to save some time)
08:10 chris yeah its the truncate ones that are having troubles
08:10 chris can you jump into mysql
08:10 magnusenger yep! :-)
08:10 chris run
08:11 chris select count(*) from zebraqueue;
08:11 chris and then the same for biblio, items, and biblioitems
08:12 paul_p magnusenger: the quick & easy way : do a truncate items, then truncate biblioitems then truncate items
08:13 chris thats what the script is trying to do paul
08:13 paul_p that should work (& much quicker than DELETE)
08:13 paul_p chris: if i'm not mistaken, not exactly.
08:13 chris thats what bulkmarcimport does
08:13 chris $dbh->do("truncate biblio");                                                                                                                                           $dbh->do("truncate biblioitems");
08:13 chris unless it has changed between 3.0.1 and 3.0.2
08:14 paul_p that's what I says, not the same (look at the order i've suggested)
08:14 paul_p if you truncate biblio 1st, then, FK constraint cascade to truncate biblioitems & items
08:14 paul_p if you truncate items 1st => no constraints
08:14 paul_p then biblioitems => no more contraints
08:14 chris yep, but we have turned FK off
08:15 chris $dbh->do("SET FOREIGN_KEY_CHECKS = 0");
08:15 paul_p magnusenger: did you set fk_off ?
08:15 paul_p (parameter on command line)
08:16 magnusenger chris: zebraque: 10, biblio: 953715, items: 0, biblioitems [still counting]. I ran with -d so i would have expected biblio, items and biblioitems to be empty...
08:16 chris yeah the truncates failed
08:16 chris due to that error
08:16 magnusenger paul_p: not actively, no
08:16 chris about locking tables
08:16 chris its because INNODB does row level locking
08:16 paul_p so that's a path to try
08:16 magnusenger chris: biblioitems: 953703
08:16 paul_p -fk
08:17 chris and if that doesnt work
08:17 paul_p wow, almost 1million biblioitems...
08:17 chris then what we do
08:17 chris is lock table biblioitems
08:17 chris truncate biblioitems
08:17 chris unlock table biblioitems
08:17 chris that will stop it doing row level locking
08:17 magnusenger paul_p: should be 400K, but they have multiplied during my attempts...
08:17 paul_p of course...
08:18 chris and you shouldnt run out of space doing the truncate, so it actually completes
08:20 magnusenger chris: sorry, lost the thread a bit. should i do truncate items, truncate biblioitems, truncate items?
08:20 chris try that
08:21 chris if it still complains about the locks
08:21 chris then we can try lock table command before each truncate
08:22 magnusenger chris: ok! thanks!
08:23 magnusenger chris: sorry that was 2x items. swap the first or last one for biblio?
08:24 chris last one biblio
08:24 magnusenger ok
08:39 chris any luck?
08:41 magnusenger truncate biblioitems is still running. guess that's a bad sign?
08:41 chris it does take a while
08:41 chris you can do
08:43 chris in another mysql console
08:43 chris show innodb status
08:44 chris the interesting bit to look at is
08:44 chris the transactions section
08:45 chris (before you run it, you might want to do)
08:45 chris pager lessl
08:45 chris whoops
08:45 chris pager less;
08:45 magnusenger show innodb status\G worked for me
08:45 chris cool
08:46 chris you should be able to see how many rows the query has affected
08:46 chris to get some idea how long it is going to take
08:47 chris if it does finish successfully
08:47 chris i would try
08:47 chris lock table biblio;
08:47 chris truncate biblio;
08:47 chris unlock table biblio;
08:47 chris for the next one
08:47 chris i suspect it will be significantly faster
08:48 chris but i have been wrong plenty of times before :-)
08:49 magnusenger ;-)
08:49 magnusenger the transactions bit looks like this does that look ok?
08:50 chris ohh someone is trying to use koha at the same time :)
08:50 chris (those are what those selects are)
08:51 magnusenger ok
08:51 chris (all queued up behind the truncate)
08:51 chris yeah its running ok, just slowly
08:52 magnusenger how can you tell the speed?
08:52 magnusenger (turned on OpacMaintenance now...)
08:52 chris show processlist;
08:52 richard cyas
08:53 chris id expect this query to take a few minutes
09:14 chris good evening jo :)
11:04 magnusenger chris: thanks for all your help so far. I gotta go, but i'll be back tomorrow...
11:53 |Lupin| does koha keep track of the changes done to a bibliographic record ?

| Channels | #koha index | Today | | Search | Google Search | Plain-Text | plain, newest first | summary