Skip Menu |

This queue is for tickets about the Schedule-Cron CPAN distribution.

Report information
The Basics
Id: 24712
Status: resolved
Priority: 0/
Queue: Schedule-Cron

People
Owner: roland [...] cpan.org
Requestors: develop [...] traveljury.com
Cc:
AdminCc:

Bug Information
Severity: Normal
Broken in: 0.97
Fixed in: (no value)



Subject: No time found
Hi Roland This is not a very useful bug report I'm afraid. I have written a daemon based on Schedule::Cron which I'm running on SuSE linux, and very very occasionally, I get an error message : 'No time found' and the daemon dies. I think it happens when system load is high, but i can't be sure. I don't know how to debug this further. What information would be useful to you? thanks Clint
Subject: No time found (Bug report for Schedule::Cron from 2007)
Hi, sorry for the huge delay (two years, or so), I was really far away from Perl living in a Java world. So, I came back to this module only recently and I'm preparing Cron::Schedule for the next release (which then probably will last again for two years untouched). So, my question is, if you still have this problem or whether I can close this ticket. If you still have an issue with Schedule::Cron I would need some more input from you. thanx, and sorry again ... ...roland
Hi Roland Show quoted text
> sorry for the huge delay (two years, or so),
no worries - I know how it goes :) Show quoted text
> I was really far away from Perl living in a Java world.
I'm not sure whether to feel sorry for you or to slap you round the head! Show quoted text
> So, my question is, if you still have this problem or whether I can > close this ticket. If you still have an issue with Schedule::Cron I > would need some more input from you.
I've been running a cron daemon based on Schedule::Cron for over 2 years now, and I've seen this issue once in a while, and have never been able to replicate it. As I said originally, it seems to happen when the system is under load - I've seen it once or twice when I was shutting the box down. I'm assuming it is some sort of race condition. I've considered putting a check into that code with a stacktrace, but it happens so seldom (like once every 6 months) that I haven't gotten around to it. I realise that this is little more than hand waving at a bug somewhere-over-there, but I hope it at least points you in the right direction. If I can give you any more info, I'd be happy to thanks clint
Hi Clinton, the error "No time found" occurs either, when no suitable next execution time could be calculated (which has not yet been occured) or when the job queue is empty, which should not be the case when you have added at least one crontab entry (because there is always a 'next' date for a cron job). I will add appropriate error messages and I'm about to release a 0.97_01 (followed by a 0.98 soon) and I also try to start some load testing to get to this error. If you like, you can consider upgrading with 0.98, which will be a drop in replacement for 0.97 with some bug fixes and an enhanced test suite. ... roland
On Tue Mar 24 06:00:45 2009, ROLAND wrote: Show quoted text
> Hi Clinton, > > the error "No time found" occurs either, when no suitable next execution > time could be calculated (which has not yet been occured) or when the > job queue is empty, which should not be the case when you have added at > least one crontab entry (because there is always a 'next' date for a > cron job).
I add all my jobs when i start the daemon, and never change the list, so yes, I should never get this error. And the fact that it happens under load makes me think that this is a timing issue: the "actual" time is later than the time it thinks it is, so jobs that should have been executed are skipped over ~~ something like that. After much hunting, I've just found the other error I have seen occasionally - just happens out of the blue, and causes the daemon to die: Can't use an undefined value as an ARRAY reference at /usr/lib/perl5/site_perl/5.8.8/Schedule/Cron.pm line 974. Show quoted text
> > I will add appropriate error messages and I'm about to release a 0.97_01 > (followed by a 0.98 soon) and I also try to start some load testing to > get to this error. > > If you like, you can consider upgrading with 0.98, which will be a drop > in replacement for 0.97 with some bug fixes and an enhanced test suite.
great - i'll be happy to. also, if you'd like to have a look at the code I've wrapped around your module, I'd be happy to show it to you. thanks for getting back on the case :) clint Show quoted text
> > ... roland
Hi Clint, this second error is even more strange than the first, since I really have no clue how this could happen since these arrays are always initialized. I added additional (probably a bit lengthy) debug for this case to see, what's going on. I'm afraid we have to wait for the bug to happen again (with 0.98 then). I will mail you, when 0.98 is out, targeted for next week. And yes, I'm interested in your wrapping code, maybe I can steal some ideas for Schedule::Cron 1.5 ;-) My E-Mail is roland@cpan.org. bye ... ...roland
Sorry, I dont know, whether this ticket still applies to the current version 1.00

If not, I would like to close this one.

Thanx ...

...roland.
CC: develop [...] traveljury.com
Subject: Re: [rt.cpan.org #24712] No time found
Date: Sat, 15 May 2010 00:57:18 +0200
To: bug-Schedule-Cron [...] rt.cpan.org
From: Clinton Gormley <clint [...] traveljury.com>
On Fri, 2010-05-14 at 13:36 -0400, Roland Huss via RT wrote: Show quoted text
> <URL: https://rt.cpan.org/Ticket/Display.html?id=24712 > > > Sorry, I dont know, whether this ticket still applies to the current version > 1.00 > > If not, I would like to close this one.
This was probably a manifestation of the same bug as deleting the PID from a hash, but in an earlier version of Perl, which didn't detect the unreferenced scalar. Hasn't happened to me for a long time - I'd close it clint Show quoted text
> > Thanx ... > > ...roland. >
Probably fixed in 1.00.
If this bug should happen in the future, a new ticket should be opened..