IRC log of #zope3-dev for Tuesday, 2006-09-19

*** flox_ has joined #zope3-dev00:09
*** RaFromBRC has quit IRC00:10
*** RaFromBRC_ has joined #zope3-dev00:10
*** whit has quit IRC00:10
*** whit has joined #zope3-dev00:10
*** philiKON has left #zope3-dev00:11
*** philiKON has joined #zope3-dev00:11
febbhello philiKON.00:12
*** d2m has joined #zope3-dev00:12
*** flox has quit IRC00:14
*** flox___ has joined #zope3-dev00:16
*** flox___ is now known as flox00:17
*** flox has quit IRC00:20
*** flox___ has joined #zope3-dev00:20
*** flox_ has quit IRC00:21
*** flox___ is now known as flox00:21
*** flox has quit IRC00:24
*** flox___ has joined #zope3-dev00:25
*** flox___ is now known as flox00:26
*** natea is now known as natea|away00:27
*** flox has quit IRC00:28
*** flox___ has joined #zope3-dev00:28
*** flox___ has quit IRC00:29
*** flox___ has joined #zope3-dev00:29
*** flox___ is now known as flox00:30
*** replicant has joined #zope3-dev00:30
*** natea|away is now known as natea00:36
*** flox has quit IRC00:43
*** gumpa_afk has left #zope3-dev00:46
*** benji has joined #zope3-dev00:56
*** replicant has quit IRC01:00
*** torkel_ has joined #zope3-dev01:05
*** J1m has quit IRC01:07
*** batlogg has quit IRC01:10
*** whit_ has joined #zope3-dev01:13
*** whit_ has quit IRC01:13
*** natea is now known as natea|busy01:14
*** whit_ has joined #zope3-dev01:14
*** ofer has joined #zope3-dev01:17
*** whit has quit IRC01:19
*** flox___ has joined #zope3-dev01:24
*** flox___ is now known as flox01:24
*** flox___ has joined #zope3-dev01:25
*** flox has quit IRC01:28
*** dunny has quit IRC01:29
*** timte has quit IRC01:41
*** replicant has joined #zope3-dev01:44
*** replicant has quit IRC01:47
*** dunny has joined #zope3-dev01:48
*** replicant has joined #zope3-dev01:49
*** RaFromBRC_ has quit IRC01:50
*** replicant has quit IRC01:53
*** niemeyer has quit IRC01:59
*** replicant has joined #zope3-dev01:59
*** whit_ has quit IRC02:03
*** whit has joined #zope3-dev02:04
*** replicant has quit IRC02:22
*** rocky is now known as rocky|Zzz02:23
*** replicant has joined #zope3-dev02:24
*** jinty has joined #zope3-dev02:25
WebMavenjinty: Hi Brian.02:27
jintyhey WebMaven02:27
* jinty is still downloading his mail;)02:28
WebMavenjinty: So, is there anything I can do to help you diagnose or test for the mysqldbda timeout problem?02:28
jintyif I make a patch, will you test it? basically that's it02:28
WebMavenjinty: Ah, OK. I'm here for the next few hours.02:28
WebMavenjinty: Yes, definitely.02:29
WebMavenBut I won't know if it worked until 8-9 hours later.02:29
jintybecause we're not in unit/functional testing kansas anymore02:29
jintyI don't mind:)02:30
WebMavenOK, then.02:30
*** ofer has quit IRC02:30
WebMavenjinty: Well, in theory, you could control MySQL's timeout with a command-line arg.02:31
WebMavenso that it times out in 1 minute.02:31
jintyand you re-connect every request;)02:32
WebMavenBut in my case, although I *think* I can do that on my desktop, in the host env MySQL isn't under my control.02:32
jintybut really you have a ready made test environment which would cost me a lot of work to duplicate.02:33
*** replicant has quit IRC02:33
*** hazmat_ has joined #zope3-dev02:33
WebMavenjinty: As I said, if you don't mind waiting, I will test it.02:33
*** yota has quit IRC02:35
WebMavenif you want to put it on a branch, I can do a checkout of the branch. If you want me to apply a patch, I can do that too.02:36
WebMavenis your change just the added code in the thread (ie. here: http://mail.zope.org/pipermail/zope3-dev/2005-December/017152.html) ?02:39
WebMavenjinty ?02:41
jintyI'm having a read thorugh the thread to remember all the details;)02:42
WebMavenAh, OK.02:42
jintyI seem to remember thinking that patch wasn't good enough02:42
jintyI think I posted a message about it, but can't remember now02:42
jintyhmm02:42
WebMavenI considered just applying it myself, but something about it made me feel... itchy.02:43
WebMavenSo, I posted to the list instead.02:43
*** whit has quit IRC02:43
*** benji has quit IRC02:45
jintywell, there's that bare except for example;)02:51
jintyWebMaven: any ideas what exception should be caught there?02:51
WebMavenUmm.02:52
WebMavenDoes the one that I'm seeing in the traceback make sense?02:52
WebMavenor is that no good because itraised after the connection has02:53
WebMavenSorry, I meant: or is that no good because it is raised after the connection has already failed?02:53
jintyno, the one ping should raise if it cannot find a connection02:55
WebMavenOh, you meant what exception should be raised... Hmm...02:57
WebMavenNo idea.02:58
jintySeems like it's this in my test:02:58
jintyTraceback (most recent call last):02:58
jinty  File "<stdin>", line 1, in ?02:58
jinty_mysql_exceptions.OperationalError: (2002, "Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2)")02:58
WebMavenWould it be the same as if you tried to make a connection to a non-existen server?02:59
jintydunno, my test was with my MySQL shut down03:04
*** dokai has quit IRC03:05
jintywhen I re-start, it magically re-connects...03:07
WebMavenSounds good.03:12
*** dokai has joined #zope3-dev03:12
WebMavenReading through the errors defined in MySQL-Python's _mysql_exeptions.py, it seems as though OperationalError is the right one.03:13
jintyok, just sent my best effort at a patch to the list03:14
jintycatches only OperationalError03:15
jintyand puts a bare except around the closing of the connection03:15
* WebMaven scratches head.03:18
WebMavenOK.03:18
*** markup_ has quit IRC03:19
jintyif close() fails, _v_connection is not set to None, and you get to start the cycle again in the next request;)03:21
*** TresEquis has joined #zope3-dev03:27
WebMavenjinty: It's not working for me.03:28
WebMavenjinty: I'm trying to paste it, but the pastebin is slow.03:31
jintyWebMaven: I also sent an alternative that's much simpler03:32
WebMavenjinty: http://paste.plone.org/1070703:32
jintyah03:33
jintyhmm well there's the reason people felt the need for the bare except03:35
WebMaventhe simpler patch works.03:37
jintywell;)03:37
jintythere's a lot to be said for simplicity03:37
WebMavenAlways.03:37
jintynow, does it work after 8 hours03:38
WebMavenI'll let you know tomorrow. ;-)03:38
WebMavenjinty: So, the first patch failed because the connection was None, so you couldn't ping it.03:45
jintyexactly03:45
WebMavenBut even if the the connection is not None, couldn't an OperationalError be raised?03:46
* WebMaven is groping in the dark here...03:47
jintyI think ping re-connects automatically03:47
jintyso if OperationalError is raised, your server is probably down03:47
WebMavenAh.03:48
jintyand then OperationalError is the least of your problems03:48
WebMavenOK.03:48
jintyor at least my experiments/the docs say ping re-connects03:48
*** wrobel has quit IRC03:48
WebMavenjinty: OK, that makes sense.03:49
WebMavenAre you going to post the simpler patch too?03:49
jintyyeah, if it works;) (maybe posting many non-functional patches to a list is a bad idea)03:52
jintyhttp://lazutkin.com/blog/2005/sep/4/django_and_mysqldb_worries/03:53
WebMavenWell, OK, but the simple patch at least doesn't make Zope barf.04:04
WebMavenWheras the first one definitely did.04:05
WebMavenHeh. "Doesn't work in an interesting way!"04:07
jintywell, the way it didn't work kinda lead me to the second patch...04:11
WebMavenalso, see this followup on the link you pasted: http://lazutkin.com/blog/2005/sep/5/django_mysql_problem_solved/04:11
jintyhmm, different problem to ours04:12
jinty_v_connection is actually a property that gets the connection from a thread local...04:14
*** alga has quit IRC04:16
*** jinty has quit IRC04:18
WebMavenDoes anyone have a generic event content type?04:27
WebMavenI mean something like a Plone Event.04:40
*** stub has joined #zope3-dev04:42
*** hazmat_ has quit IRC05:15
*** replicant has joined #zope3-dev05:44
febbhello again06:03
*** roym has quit IRC06:21
*** baijum has joined #zope3-dev06:35
*** TresEquis has quit IRC07:21
febbhi all:   any ideas regarding this error after passing all tests (make check)  , this is python Fedora Core 5, Python 2.4.3, Zope 3.2.1:07:35
febb  Ran 8045 tests with 0 failures and 0 errors in 1019.707 seconds.07:35
febbError in atexit._run_exitfuncs:07:35
febbTraceback (most recent call last):07:35
febb  File "/usr/lib/python2.4/atexit.py", line 24, in _run_exitfuncs07:35
febb    func(*targs, **kargs)07:36
febb  File "/usr/lib/python2.4/logging/__init__.py", line 1332, in shutdown07:36
febb    h.flush()07:36
febb  File "/usr/lib/python2.4/logging/__init__.py", line 718, in flush07:36
febb    self.stream.flush()07:36
febbValueError: I/O operation on closed file07:36
*** timte has joined #zope3-dev08:02
*** eins has joined #zope3-dev08:07
einshi all08:07
febbhi eins08:07
einsfebb hey08:12
eins;)08:12
febbeins, how are you  ?08:12
einsprobably quite well ;)08:18
* eins has catched a cold08:18
*** romanofski has quit IRC08:21
febbthats too bad.  hope you get well.    I am trying to solve an installation error with Z3, just after the make check and sucessful tests.08:25
febbthen I get a  ValueError: I/O operation on closed file  and traceback....08:26
*** dobee has joined #zope3-dev08:47
*** kobold has joined #zope3-dev08:52
*** alecm has quit IRC09:06
*** BjornT has quit IRC09:14
*** hdima has joined #zope3-dev09:14
*** BjornT has joined #zope3-dev09:20
*** romanofski has joined #zope3-dev09:32
*** batlogg has joined #zope3-dev09:32
romanofskimoin09:33
* baijum wathing Keynote by Martijn Faassen http://podcasts.plonetv.de/ep05/cltv05-mpeg.mov09:37
* baijum Zope History is really interesting, Zope 3 is great but not zope.org..09:39
*** kobol1 has joined #zope3-dev09:44
* baijum : Yes, we need to actively engage our Python friends/competitors09:44
*** kobold has quit IRC09:46
*** kobol1 is now known as kobold09:46
*** replicant has quit IRC09:56
*** flox has joined #zope3-dev10:13
*** philiKON has quit IRC10:15
*** MJ has quit IRC10:27
*** yota has joined #zope3-dev10:29
*** BjornT has quit IRC10:36
*** BjornT has joined #zope3-dev10:46
*** runyaga_ has joined #zope3-dev10:51
*** runyaga__ has joined #zope3-dev10:54
*** srichter has joined #zope3-dev10:55
*** ChanServ sets mode: +o srichter10:55
*** philiKON has joined #zope3-dev10:59
*** runyaga_ has quit IRC11:03
*** runyaga has quit IRC11:09
*** regebro has joined #zope3-dev11:11
*** MJ has joined #zope3-dev11:13
*** replicant has joined #zope3-dev11:24
*** timte has quit IRC11:29
*** timte has joined #zope3-dev11:30
*** BjornT has quit IRC11:32
*** projekt01 has joined #zope3-dev11:33
*** replicant has quit IRC11:36
*** BjornT has joined #zope3-dev11:37
*** volvox has joined #zope3-dev11:43
*** regebro has quit IRC11:52
*** jinty has joined #zope3-dev12:07
*** mgedmin has joined #zope3-dev12:47
*** regebro has joined #zope3-dev12:59
*** WebMaven has quit IRC13:01
*** WebMaven has joined #zope3-dev13:05
*** baijum_ has joined #zope3-dev13:06
*** baijum has quit IRC13:07
*** projekt01 has left #zope3-dev13:11
*** dunny has quit IRC13:17
*** baijum_ has quit IRC13:27
*** mkerrin has joined #zope3-dev13:30
*** rocky|Zzz is now known as rocky13:33
*** mkerrin has quit IRC13:46
*** mkerrin has joined #zope3-dev13:56
*** baijum has joined #zope3-dev13:59
*** J1m has joined #zope3-dev14:02
*** BjornT has quit IRC14:03
*** BjornT has joined #zope3-dev14:07
*** andrew_m has left #zope3-dev14:21
*** alga has joined #zope3-dev14:48
*** ignas has joined #zope3-dev14:48
*** vlado has joined #zope3-dev14:53
*** nathany has joined #zope3-dev15:28
*** zagy_ has quit IRC15:34
*** niemeyer has joined #zope3-dev15:36
*** tfws has joined #zope3-dev15:50
*** baijum has left #zope3-dev15:51
*** baijum has joined #zope3-dev15:52
*** MJ has left #zope3-dev15:52
baijumhttp://digg.com/programming/Python_2_5_final_released15:57
*** baijum has quit IRC16:17
*** Aiste has joined #zope3-dev16:19
*** whit has joined #zope3-dev16:21
*** alecm has joined #zope3-dev16:23
faassenwhit: hey16:40
whithiya faassen16:40
faassenwhit: just sent you another email.16:41
whitcool16:41
* whit checks16:41
*** nathany has quit IRC17:00
*** tfws has quit IRC17:03
*** nathany has joined #zope3-dev17:03
*** eins has quit IRC17:06
*** RaFromBRC has joined #zope3-dev17:09
*** whit has quit IRC17:13
*** whit has joined #zope3-dev17:18
*** volvox has quit IRC17:22
*** regebro has quit IRC17:35
*** hdima has quit IRC17:38
*** MJ has joined #zope3-dev17:42
*** batlogg has quit IRC17:54
WebMavenjinty: AYT?18:00
jintyhey WebMaven18:00
jintyany luck?18:00
WebMavenDidn't work.18:00
jintyhmm, can you pastebin something?18:01
WebMavenYes. One moment.18:01
mgedminpastey.net is faster than pastebin.com18:01
WebMavenhttp://paste.plone.org/1071118:03
WebMavenjinty: so, any ideas?18:07
* jinty curses the docstring of the ping function18:08
* WebMaven summons the forces of darkness and chaos to assist in jinty's curse...18:09
WebMavenjinty, I'll be afk for about 10 minutes, OK?18:12
jintysure18:13
jintybusy mixing the two patches from yesterday18:13
WebMavenjinty: back18:23
jintyhmm, if ping fails, do we really need to close the connection?18:25
WebMavendunno.18:28
WebMavenWasn't ping supposed to automatically reopen the connection?18:29
*** flox has quit IRC18:30
jintyyeah, it was18:30
WebMavenLooking at the 'Jim Washington' patch again...18:30
WebMavenOK, so your objection to that one was mostly the bare except, right?18:31
jintymysql_ping will only attempt to automatically reconnect if that option has been previously set. As of 5.0.3 the default is to not reconnect.18:31
jintydo you have mysql greater than 5.0.3?18:32
WebMavenLet me check...18:32
jintyfrom here: http://dev.mysql.com/doc/refman/5.0/en/mysql-ping.html18:32
jintyanyway, sent you a new patch18:32
*** ktwilight has quit IRC18:32
WebMaven 5.0.1918:33
*** natea|busy is now known as natea18:34
*** dokai has quit IRC18:37
WebMavenjinty: still haven't got it...18:37
jintyresent18:42
WebMavenok, got it this time.18:43
WebMavenHmm.18:43
WebMavenIs the retry stuff necessary?18:43
WebMavenDoes that result in a prgressive backing off behavior?18:43
WebMavenAlso, would it be cleaner if you were testing for 'is None', and returning False, and then put everything else in an 'else'?18:46
jintywell, I presume the publisher will only retry the request a certain number of times.18:46
*** vlado has quit IRC18:46
jintyfeel free to modify:)18:47
jintyyeah it does look nicer18:49
*** markup_ has joined #zope3-dev18:49
*** ktwilight has joined #zope3-dev18:50
WebMavenjinty: Will the publisher retrying ever result in the connection being re-established?18:53
jintyyes, precisely becuse we set _v_connection to None18:54
jintyto on the next retry, isConnected returns false18:54
jintyand then it's caller connect() in zope.rdb will connect.18:54
WebMavenAha. So it's an indirect recursion.18:54
jintythis way you can do the right thing even if you lose your connection half way through the request18:55
jintyonce you've already sent SQL to the database18:55
WebMavenHmm.18:55
WebMavenOK.18:55
*** benji has joined #zope3-dev18:56
WebMavenBecause just returning False will lose that info?18:56
jintyyep.18:57
* jinty get's very worried about persistent DA's now18:57
WebMavenjinty: does this look better: http://pastey.net/93518:57
jintyyeah;)... I would even get rid of the else19:00
WebMavencheck it now: http://pastey.net/93519:03
*** stub has quit IRC19:05
*** stub has joined #zope3-dev19:05
jintyyep19:05
*** MJ has quit IRC19:05
WebMavenWell, it works right now. We'll see whether it deals correctly with a timeout19:08
WebMavenDo you need me to email you my revision?19:08
WebMavenjinty: Do you need me to email you my revision?19:11
jintyif it works;)19:14
*** jhauser has joined #zope3-dev19:16
febbHello all.19:17
febbAre there known bugs in Zope 3 (3.2.1) regarding the use of the python (2.4.3) logging mechanism ? (As I am getting a consistent I/O error operation on closed file just after the 'make check' even tough all tests passed sucessfully).19:18
*** dobee has quit IRC19:19
WebMavenjinty: emailed.19:21
jintyWebMaven, thanks, then I guess till tomorrow19:22
srichterfebb: not that I know19:23
WebMaventomorrow, then. Cheers.19:23
WebMavenjinty: uh, wait...19:23
WebMavenI just noticed that your patch was against a later revision...19:24
*** dobee has joined #zope3-dev19:24
WebMavenyou're on rev 70218, My checkout is 7016319:24
*** flox has joined #zope3-dev19:25
WebMavenOh, nm.19:25
WebMavenI just did an svn up, with no changes.19:26
febbsritcher, I am getting that error consistently.... let me put the output on rafb.net/paste...19:26
febbhere it is :  http://rafb.net/paste/results/yMiue198.html19:30
*** philiKON has quit IRC19:33
*** replicant has joined #zope3-dev19:34
*** RaFromBRC is now known as RaFromBRC|away19:44
*** dobee has quit IRC19:44
*** jhauser has quit IRC19:45
*** whit has quit IRC19:57
*** faassen has quit IRC20:02
*** batlogg has joined #zope3-dev20:19
*** batlogg has quit IRC20:26
*** batlogg has joined #zope3-dev20:27
*** philiKON has joined #zope3-dev20:28
*** ignas has quit IRC20:33
*** kobold has quit IRC20:48
*** flox has quit IRC20:55
*** whit has joined #zope3-dev20:58
*** flox has joined #zope3-dev21:03
*** dokai has joined #zope3-dev21:10
*** whit has quit IRC21:27
*** dunny has joined #zope3-dev21:36
*** srichter has quit IRC21:39
*** zbir has quit IRC21:41
*** whit has joined #zope3-dev21:42
*** dobee has joined #zope3-dev21:51
*** MJ has joined #zope3-dev21:54
*** RaFromBRC|away is now known as RaFromBRC21:57
*** alecm has quit IRC22:00
*** whit has quit IRC22:03
*** zbir has joined #zope3-dev22:04
*** whit has joined #zope3-dev22:06
*** gumpa has joined #zope3-dev22:06
*** GaryPoster has joined #zope3-dev22:09
*** mkerrin has quit IRC22:10
*** reco has joined #zope3-dev22:14
*** GaryPoster has quit IRC22:14
*** whit has quit IRC22:18
*** whit has joined #zope3-dev22:19
*** dunny has quit IRC22:24
philiKONJ1m, have you seen martijn's keynote at the DZUG conf yet?22:28
J1mYup22:29
J1mPretty funny22:29
J1mand lots of good points.22:29
philiKONyep22:29
*** dobee has quit IRC22:37
*** srichter has joined #zope3-dev22:44
*** ChanServ sets mode: +o srichter22:46
*** philiKON has quit IRC22:47
*** mgedmin has quit IRC22:48
*** philiKON has joined #zope3-dev22:48
*** whit has quit IRC22:58
*** fcorrea has joined #zope3-dev23:06
*** romanofs1i has joined #zope3-dev23:07
*** romanofs1i is now known as romanofski_23:07
*** whit has joined #zope3-dev23:13
*** flox has quit IRC23:19
*** flox has joined #zope3-dev23:20
*** RaFromBRC has quit IRC23:20
*** whit has quit IRC23:22
*** stub has quit IRC23:25
*** alecm has joined #zope3-dev23:40
*** whit has joined #zope3-dev23:44
*** flox_ has joined #zope3-dev23:51
*** vlado has joined #zope3-dev23:54
*** flox has quit IRC23:57

Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!