Today | Next day → | Search | Index
All times shown according to UTC.
| Time | Nick | Message |
|---|---|---|
| 19:29 | kados | not there quite yet owen ;-) |
| 19:30 | owen | What's the latest? |
| 19:31 | kados | still running updatedatabase |
| 19:32 | owen | Wow...I guess it always takes longer than we expect |
| 19:33 | shedges | a..l..w..a..y..s.. |
| 19:34 | kados | yep ... I'm bad at estimations |
| 19:38 | owen | I've been able to stop by, but I've been pretty busy with family stuff all day, so I'm glad we didn't have a bunch of urgent stuff going on |
| 19:56 | rach | hello |
| 19:56 | owen | Hi rach |
| 19:57 | rach | hi owen |
| 19:57 | owen | We're in the midst of a 2.2 upgrade today |
| 19:57 | rach | oh wow |
| 19:58 | owen | The database has been updating for about 12 hours now |
| 20:28 | shedges | Hi rachel! |
| 20:33 | kados | so ... updatedatabase just finished |
| 20:33 | rach | hi stephen |
| 20:34 | shedges | kados: how much longer? |
| 20:36 | kados | shedges: updatedatabase is finished ... ready for testing? |
| 20:36 | shedges | yep. I'll start with OPAC(?) |
| 20:36 | kados | ok ... |
| 20:36 | shedges | (what's the end of the IP?) |
| 20:37 | kados | it's https://66.213.78.69 |
| 20:37 | shedges | refused... |
| 20:37 | kados | what's your IP |
| 20:37 | shedges | ah httpS |
| 20:37 | kados | well ... I'll need to give you access if you're at home |
| 20:38 | what does it tell you when you go to whatismyip.com | |
| 20:39 | shedges | 65.24.138.126 |
| 20:41 | ok, thanks! | |
| 20:41 | kados | ok ... opac is at http://66.213.78.67 and intranet is at https://66.213.78.69 |
| 20:41 | I'm gonna head to APL ... BRB | |
| 20:46 | shedges | opac doesn't automaticallt insert the zeroes in the card number anymore |
| 20:47 | (I think Owen will need to apply Oscar's java script) | |
| 20:52 | kados | yep ... |
| 20:52 | shedges | you're back! |
| 20:53 | kados | yep just got to APL |
| 20:53 | shedges | Another owen issue -- confusion over the copyright "or" recent acqusition parameters (or only applies to those two, not the other stuff on that page) |
| 20:53 | kados | owen has some configuration to do in parameters too iirc |
| 20:55 | so as far as custom scripts to I've been putting them in /build/utilities/ | |
| 20:56 | shedges | overduenotices.pl ? |
| 20:57 | kados | yea ... I haven't uploaded the latest copy of that ... is it on .76? |
| 20:57 | shedges | yep |
| 20:57 | i'm in intranet, now -- nice! | |
| 20:57 | kados | yea ... looks great |
| 21:00 | shedges | wonder if the self checkout will work... |
| 21:02 | kados | yep ... already does |
| 21:03 | that virtualhost isn't online atm | |
| 21:03 | but I tested it | |
| 21:03 | so I checked .76 cron and it looks like the only item we run in cron is reservelist.pl ... is that right? | |
| 21:03 | shedges | yep -- the only one I know of anyway. |
| 21:04 | kados | ok ... |
| 21:04 | that's setup on the new server | |
| 21:05 | shedges | (I REALLY like this Intranet!) |
| 21:05 | kados | it's SSLed too ;-) |
| 21:08 | so my next step would be to pull out the ethernet cords on 77 and 76, configure 69 with those IPs and make sure all our DNS is working the way we expect before I reboot the machines and swap RAM | |
| 21:09 | shedges | sounds good |
| 21:09 | kados | ok ... |
| 21:09 | shedges | I'm going to ask the catalogers to wait until I get there to do their import tomorrow. |
| 21:14 | uh -oh The book bag doesn't work in the OPAC. (When I choose items to add from the serach results screen, it adds entirely different items -- the right number of items, but the wrong items) | |
| 21:19 | hi owen | |
| 21:19 | owen | Hi |
| 21:19 | shedges | we're testing |
| 21:20 | owen | updatedatabase finally finished? |
| 21:20 | shedges | yep, and Joshua's in the process of switching machines (I think) |
| 21:20 | owen | What can I do? |
| 21:20 | shedges | found a few things for your list |
| 21:20 | opac doesn't automatically insert the zeroes in the card number anymore (I think Owen will need to apply Oscar's java script) | |
| 21:21 | Another owen issue -- confusion over the copyright "or" recent acqusition parameters (or only applies to those two, not the other stuff on that page) | |
| 21:21 | The book bag doesn't work in the OPAC. (When I choose items to add from the search results screen, it adds entirely different items -- the right number of items, but the wrong items) | |
| 21:21 | but generally -- you guys DONE GOOD!! | |
| 21:22 | IE still does some funky stuff, too, but I expect that anymore... | |
| 21:22 | kados | hmmm ... I'll have to check out that book bag error later |
| 21:23 | owen | Must be something in the template. |
| 21:23 | The the opac still at .67? | |
| 21:23 | shedges | can owen see -- is the switch-over done? |
| 21:25 | kados | yea ... |
| 21:25 | owen what's your IP at home? | |
| 21:25 | no the switch isn't done | |
| 21:25 | but he can see the opac (not the intranet tho) | |
| 21:26 | owen | Is there an easy way to find out from the Win2K command prompt? |
| 21:26 | shedges | whatismyip.com |
| 21:31 | (that's a website, sorry -- I guess it could be a W2K command!) | |
| 21:31 | owen | Worked fine. |
| 21:31 | ...but my opac search timed out. | |
| 21:32 | shedges | mine too, that machine must be unplugged now |
| 21:33 | (time to play Solitaire for a while...) | |
| 21:41 | owen | shedges: what do you mean about the copyright "or" recent acquisition parameters? |
| 21:42 | shedges | Cheryl was trying the OPAC and wasn't sure from the page payout if she could choose branch and recent acquisitions |
| 21:43 | thought it looked a little like the whole page was "or" choices, no "ands" | |
| 21:44 | owen | Okay. I'll take another look at that. |
| 21:46 | shedges | Seeing the way you solved some thorny problems (like displaying branch availability), I'm sure you come up with something clever! |
| 21:52 | kados | update: search.athen... and intranet.athe... are working |
| 21:53 | owen did koha.athenscounty.lib.oh.us _ever_ point to the opac? | |
| 21:53 | or was that always the intranet interface? | |
| 21:53 | owen | yes |
| 21:54 | yes, it did point to the opac | |
| 21:54 | kados | ok ... |
| 21:54 | shedges | the overdue notice spoint people to koha.athen... I'll need to change that. |
| 21:54 | owen | and it pointed to the intranet with the 8080 port |
| 21:55 | Can we have the koha.athens... url still point to the same machine anyway? | |
| 21:56 | shedges | owen, I think all opac-format-search needs is a little space between "limit by branch" and "limit by copyright date" |
| 21:58 | owen | kados: if I'm sftp'ing should I still use the .67 address? .69? |
| 22:01 | kados | owen: no ... we're online with the 76 77 addresses ... 66-69 are for test servers |
| 22:03 | shedges | Do I access mysql on 77? |
| 22:07 | owen | Hmmm... little network hiccup there. |
| 22:07 | That's a strange problem with the book bag...we seem to be getting randomly incorrect biblionumbers here and there | |
| 22:08 | The first test I did showed problems, and the second didn't. The third had only one bad one. | |
| 22:08 | shedges | were they recent adds? old biblios? |
| 22:09 | owen | I did a keyword search for 'scrapbook,' and one of the results was 'Ben Franklin's almanac' |
| 22:10 | shedges | exit |
| 22:10 | owen | The link to the item has the correct biblionumber, but the id attached to the checkbox (for the bookbag) is different |
| 22:11 | I'll bet it's a biblionumber/biblioitemnumber thing, and I've got the wrong variable in the template. | |
| 22:13 | shedges | seems like it should _never_ work, if that is the case |
| 22:13 | kados | shedges: you can access mysql on either IP |
| 22:14 | shedges | looks like it has a new password, i'll call sometime and get that |
| 22:16 | owen | Yep, it was using the biblioitemnumber instead of the biblionumber. And it just happened to work most of the time. I wonder if that bug has been around since we first upgraded the opac? |
| 22:16 | kados | kohaadmin should be the same ... |
| 22:17 | owen did you make the changes to params ... I can't remember exactly what they were ... something to do with frameworks? | |
| 22:17 | interesting ... so is it fixed owen? | |
| 22:17 | owen | Yes |
| 22:18 | kados | cool |
| 22:18 | owen | I had set up some biblio frameworks, and I'll have to do that again but it's not urgent |
| 22:24 | kados | cool ... have a good evening |
| 22:25 | afaik the only thing left is swapping out the RAM and getting the self-checkout Vhost up | |
| 22:25 | oh ... and I need to make sure that koha.athenscounty.lib.oh.us:8080 points to the right place | |
| 22:26 | shedges | sounds good. Nice work, Joshua! |
| 22:34 | kados | hmmm ... setbranch just threw an error |
| 22:36 | shedges | looks like it's pointing to selectbranchprinter.pl |
| 22:36 | kados | ahh ... wrong permissions |
| 22:36 | shedges | works now! |
| 22:40 | "Reserve List" won't work until cron does it's thing -- right? | |
| 22:42 | kados | afaik ... |
| 22:43 | shedges | But now I see "Format Search" (intranet-item_report.php) also doesn't work. |
| 22:44 | kados | hmmm, we'll have to ask owen about that one |
| 22:44 | i wonder if it's related to the frameworks ... | |
| 22:45 | shedges | they're both php reports... |
| 22:45 | I don't think it's something to do with frameworks. | |
| 22:47 | kados | hmmm, reservlist.php isn't even in the same place is it... |
| 22:48 | shedges | intranet-reserve_list.php? |
| 22:48 | kados | hmmm, james has it bookmarked as reserve_list.php on his machine ... |
| 22:49 | shedges | confusing... |
| 22:50 | owen's biblio frameworks show up in the pull down list on the cataloguing screen, they just default right now to the full MARC framework. | |
| 22:54 | kados | ahh ... I think I know one of the probs with owens scripts ... he hardcoded the IP which has changed ... |
| 22:55 | maybe ... although I don't see that in the scripts | |
| 22:55 | shedges | well, the world won't end if we don't have a reserve list tomorrow. |
| 22:55 | kados | ahha ... here it is |
| 22:55 | you sure? ;-) | |
| 22:56 | shedges | I've beat on about everything I can think of -- it looks to me like we can all go to work tomorrow. |
| 22:57 | kados | reservelists included ;-) |
| 22:58 | shedges | yep -- nice! |
| 22:58 | what a minute -- !! Thought I saw it working. | |
| 22:59 | kados | it's not? |
| 22:59 | it's working for me | |
| 23:00 | shedges | I'm click on "Reserve List" in the left navigation bar -- takes me to the main page |
| 23:07 | kados | shedges: so if it's alright with you I'll swap RAM and optimize mysql ... |
| 23:07 | requires a reboot ;-) | |
| 23:08 | shedges | yep, I'm done for the night. Thanks, Joshua! |
| 23:22 | owen | Bookmarks for the reserve list will be incorrect now. |
| 23:22 | We'll have to post a message to the tech forum. | |
| 23:23 | Stephen said the intranet format search wasn't working? | |
| 23:27 | kados | reboots are all good |
| 00:01 | rach | joshua are you there/ |
| 00:01 | and can you answer a dumb question for me? | |
| 00:40 | kados | rach: here now |
| 00:40 | what's up? | |
| 00:41 | rach | ah I wanted to know how you might tell how many people were in a que waiting to to get items issued (without being able to see them) I assume you can't, but wondered if there was some sort of assumption based statistic, like if you get this many people issued in 5 mins then they must have been in a que |
| 00:41 | kados | right ... so you mean for reserves? |
| 00:41 | rach | but I rang HLT and they reckoned you couldn't |
| 00:41 | kados | we have a special script for that |
| 00:42 | rach | no I think they mean a physical queue |
| 00:42 | kados | ahh ... I understand |
| 00:43 | well Koha could surely provide data for that | |
| 00:43 | we'd need a statitition to tell us how to calculate it | |
| 00:43 | but it's probably possible | |
| 00:43 | (like they do in phone queues) | |
| 00:43 | rach | yep i reckong it's not really possible |
| 00:43 | without some sort of pressure matt :-) | |
| 00:43 | kados | harry seldon would beg to differ ;-) |
| 00:44 | rach | or the librarian looking up and counting them |
| 00:44 | which would seem a lot easier | |
| 01:34 | kados | paul around? |
| 05:14 | paul | salut hdl |
| 05:14 | hdl | salut paul |
| 05:15 | paul | kados, i'm here (9AM in France now, was 4AM when you "called" me, so I was in bed. I suspect you are in bed now ;-) |
| 06:37 | sylvain | hi all |
| 06:37 | paul | coucou |
| 06:45 | sylvain | tiens paul une petite question si tu as le temps, quand je crée un lecteur, pourquoi est-ce que sa date "joined" n'est pas initialisée ? Ca se spécifie quelque part ? |
| 06:46 | ok, merci quand mm ;) | |
| 07:19 | paul, une autre question du même style, dans la table borrowers, on a un champ area, tu sais à quoi il correspond ? | |
| 07:20 | paul | non pas trop. |
| 07:22 | sylvain | tu sais à qui je pourrais poser des questions concernant les lecteurs ? |
| 07:49 | une petite question encore (gros début de semaine :)) dans la fonction C4/Accounts2/manualinvoice on a des descriptions du style 'new card', 'lost item' ... Au niveau internationalisation, quand c'est pas dans des templates, il y a une méthode particluière pour les traductions ? | |
| 07:51 | paul ? | |
| 07:51 | paul | (au tel) |
| 07:51 | sylvain | ok |
| 08:44 | bon avant d'aller manger, une autre question, le sens de la colonne lost dans la table borrowers si quelqu'un sait ;) | |
| 09:12 | paul | syl_away : là je sais, c'est lorsque la carte lecteur est perdue : on peut "désactiver" le lecteur par ce biais. Si le lost est coché, alors prêt impossible. |
| 10:02 | kados | hi paul ;-) |
| 10:03 | paul | 'morning joshua |
| 10:03 | kados | our server info: 1Ghz pentium, 2gig of ram, 10,000rpm disks ... not a showstopper |
| 10:04 | paul | just with specific mysql tuning (mysql cache in RAM) |
| 10:04 | kados | last night we migrated to 2.2 |
| 10:04 | paul | 4 disks in RAID 5 ? |
| 10:04 | kados | no raid |
| 10:04 | paul | i saw you chat with stephen about updatedatabase... |
| 10:04 | ok, thanks | |
| 10:04 | just 1Ghz pentium ? | |
| 10:04 | kados | three disks, one for database, one for filesystem and one for backups, etc. |
| 10:05 | yep | |
| 10:05 | paul | ok. And how many issues per year ? |
| 10:05 | kados | our circ is 620,000 |
| 10:05 | 30,000 borrowers | |
| 10:05 | paul | how many librarian computer ? |
| 10:05 | what kind of network ? | |
| 10:05 | (10MB ? 100MB ? less ? more ?) | |
| 10:06 | i mean between branches | |
| 10:06 | kados | two in each branch for circ ... plus at least two others for other purposes (behind the desk) |
| 10:06 | our network is mostly 1.5Kb | |
| 10:06 | paul | do you have an idea of opac use ? |
| 10:06 | 1.5Kb ??? | |
| 10:06 | you mean 1500 BYTES .? | |
| 10:06 | kados | sorry ... I meant 1.5 Mb |
| 10:06 | ;-) | |
| 10:06 | paul | you must be kidding |
| 10:06 | ah, ok ;-) | |
| 10:07 | any idea of opac use ? | |
| 10:08 | kados | hmmm, well we get about 500-1000 queries per day |
| 10:09 | paul | you mean opac query or apache query ? |
| 10:09 | kados | opac queries |
| 10:09 | I wrote a small addition to SearchMarc that tracks all opac searches | |
| 10:10 | paul | useful for official release ? |
| 10:10 | kados | sure ... it goes along with the suggest and spellcheck features ... |
| 10:10 | I'll commit it asap | |
| 10:11 | paul can you run this on one of your machines for me? | |
| 10:11 | explain select distinct m1.bibid from biblio,biblioitems,marc_biblio,marc_word as m1,marc_word as m2 where biblio.biblionumber=marc_biblio.biblionumber and biblio.biblionumber=biblioitems.biblionumber and m1.bibid=marc_biblio.bibid and (m1.bibid=m2.bibid) and ((m1.word like 'book')and (m2.word like 'boys')) order by biblio.title; | |
| 10:11 | I've got some perf probs since the upgrade ... I'm afraid my indexes are not optimized: | |
| 10:11 | | m2 | range | bibid,word | word | 255 | NULL | 839 | where used; Using temporary; Using filesort | | |
| 10:11 | | m1 | ref | bibid,word | bibid | 8 | m2.bibid | 34 | where used | | |
| 10:11 | | marc_biblio | eq_ref | PRIMARY,biblionumber | PRIMARY | 8 | m1.bibid | 1 | where used; Distinct | | |
| 10:11 | | biblio | eq_ref | PRIMARY,blbnoidx | PRIMARY | 4 | marc_biblio.biblionumber | 1 | Distinct | | |
| 10:11 | | biblioitems | ref | bibnoidx | bibnoidx | 4 | biblio.biblionumber | 18 | Using index; Distinct | | |
| 10:12 | paul | I get : |
| 10:12 | table | |
| 10:12 | ||
| 10:12 | type | |
| 10:12 | ||
| 10:12 | possible_keys | |
| 10:12 | ||
| 10:12 | key | |
| 10:12 | ||
| 10:12 | key_len | |
| 10:12 | ||
| 10:12 | ref | |
| 10:12 | ||
| 10:12 | rows | |
| 10:12 | ||
| 10:12 | Extra | |
| 10:12 | mmm... stupid paste. | |
| 10:12 | kados | I wrote my own order by sub for SearchMarc so I've been able to get rid of the 'using filesort' ... |
| 10:13 | paul | I have the same explain |
| 10:13 | | m1 | range | bibid,word | word | 255 | NULL | 1 | Using where; Using temporary; Using filesort | | |
| 10:13 | | marc_biblio | eq_ref | PRIMARY,biblionumber | PRIMARY | 8 | m1.bibid | 1 | Using where; Distinct | | |
| 10:13 | | biblio | eq_ref | PRIMARY,blbnoidx | PRIMARY | 4 | marc_biblio.biblionumber | 1 | Distinct | | |
| 10:14 | | biblioitems | ref | bibnoidx | bibnoidx | 4 | biblio.biblionumber | 1 | Using index; Distinct | | |
| 10:14 | kados | but the keys don't look right (sndx_word and Search_Marc aren't working?) |
| 10:14 | paul | | m2 | range | bibid,word | word | 255 | NULL | 1 | Using where; Distinct | |
| 10:14 | kados | one major difference |
| 10:14 | the rows on yours | |
| 10:14 | that's my problem ... I wonder how to solve it | |
| 10:15 | any ideas? | |
| 10:17 | paul | isn't my "row" column due to a "english" query where i have only french db ? |
| 10:18 | kados | hmmm, maybe ... how about doing a french query ... |
| 10:18 | paul | with a french query, I get results like yours : |
| 10:18 | kados | ok |
| 10:18 | paul | 47 1 1 12 20 |
| 10:18 | (that is not very good i agree) | |
| 10:19 | that's why marc_words is a must-have-asap... | |
| 10:19 | kados | hmmm, 839 is REALLY not good ... I'd take 47 happily:-) |
| 10:19 | I agree | |
| 10:19 | paul | for sure, but my DB is only 20 000 biblios |
| 10:19 | where your is 10x this ! | |
| 10:20 | kados | right |
| 10:20 | opac queries are pretty laggy (6-10 sec) ... I'm going to try to figure out if there a way to optimize further | |
| 10:20 | (given our limitations) | |
| 10:21 | paul | mmm... |
| 10:21 | strange because almost nothing has changed since 2.0 on this side of Koha | |
| 10:21 | (except order by) | |
| 10:21 | kados | that's what I was thinking |
| 10:22 | (and I took out the order by ) ;-) | |
| 10:22 | paul | (maybe a good place to investigate... remove the orderby) |
| 10:22 | ok, so, really strange... | |
| 10:22 | kados | yep .. well that gets rid of the filesort |
| 10:22 | so what are sndx_word and Search_Marc indexes for? | |
| 10:23 | paul | strange. they are supposed to be used, you're right. |
| 10:23 | sylvain | vu pour le lost paul merci |
| 10:23 | paul | maybe somethings changed but i've forgotten at the moment |
| 10:32 | kados | so the m1 doesn't use index, the marc_bibli doesn't use index and the biblio doesn't use index ... that's a lot of missing indexes for a search |
| 10:32 | (add the m2 to that list ;-)) | |
| 10:45 | so paul ... I've a few questions about your libraries: which is the biggest? and what are annual circ, # of borrowers, # of items? | |
| 10:45 | paul | lower than you ;-) |
| 10:45 | the biggest will be 100 000 items when all catalogued, for instance 18 000 | |
| 10:46 | Next week i'll put to production a library with 45 000 items, that will be my biggest | |
| 10:46 | and i've no libraries with more than 1 branch | |
| 10:47 | max circ should be something like x0 000 | |
| 10:47 | although most have only a "few" circ, because they are search libraries | |
| 10:47 | with a very specific content, and only a few readers. | |
| 10:52 | kados | I see |
Today | Next day → | Search | Index