Page 1 of 1

[0.5.3] CherryPy MemoryError

Posted: May 8th, 2010, 4:40 pm
by jocke
Version: 0.5.0 Final
OS: Win XP SP2
Install-type: Windows Installer
Skin (if applicable): smpl Version: 1.3
Firewall Software: None
Are you using IPV6? It's enabled in the OS, but not in use by SABnzbd
Is the issue reproducible? No

Haven't updated to 0.5.2 yet, and I've been away for a while. When I got back today, the following met me when opening SABnzbd;

Image

The daemon itself runs, and there are no obvious error-logs. Ended the process, started it, and it was fine. I'm pretty sure this is a one-time-thing, but thought I'd post it here, just to be sure.

Re: [0.5.0 Final] CherryPy MemoryError

Posted: May 8th, 2010, 4:49 pm
by shypike
The web server is quite independent from the rest of the software.
Also we run a patched pre-release of CherryPy, which probably contains some errors.
It's good enough for our purposes.
0.5.2 is worth the upgrade, but it contains exactly the same CherryPy.
Thanks for the report, for now I'll consider it a flaky.

Re: [0.5.0 Final] CherryPy MemoryError

Posted: October 14th, 2010, 4:24 pm
by jocke
It happened again. Exactly the same error. Same CherryPy-version (3.2.0), but this time with SABnzbd+ 0.5.3.

Re: [0.5.3] CherryPy MemoryError

Posted: October 14th, 2010, 6:52 pm
by shypike
>>> DO NOT USE CHERRYPY 3.2.0 <<<<
Use the CherryPy that's in our tar.gz distribution file.

Re: [0.5.3] CherryPy MemoryError

Posted: October 15th, 2010, 7:45 am
by jocke
Was the CherryPy-version changed between 0.5.3 and 0.5.4?

Re: [0.5.3] CherryPy MemoryError

Posted: October 15th, 2010, 8:57 am
by shypike
No, we have been using the same since 0.5.0.
0.4.x uses another one.

Re: [0.5.3] CherryPy MemoryError

Posted: October 15th, 2010, 11:00 am
by jocke
So SABnzbd-upgrades doesn't upgrade CherryPy? Because I'm running SABnzbd 0.5.3, and my CherryPy-version is still 3.2.0.

Re: [0.5.3] CherryPy MemoryError

Posted: October 15th, 2010, 6:01 pm
by shypike
To avoid confusion.
Are you using the CherryPy included in SABnzbd's distribution or
a CherryPy from your OS's packaging system?
You must use the included one.
The version number in the included one is not reliable, because it's a patched test release.

Re: [0.5.3] CherryPy MemoryError

Posted: October 15th, 2010, 8:24 pm
by jocke
shypike wrote:Are you using the CherryPy included in SABnzbd's distribution?
Yes. If you see my first post, I'm using Windows.

Re: [0.5.3] CherryPy MemoryError

Posted: October 16th, 2010, 3:14 am
by shypike
Sorry, I lost track (I have to monitor too many track).
I was convinced you were experimenting with different CherryPy modules.

It seems you simply hit a bug in CherryPy.
What happens inside CherryPy is a black box to use, we just use it.
Does this issue result in any loss of functionality?
Does the UI recover when you just refresh the browser page?

Re: [0.5.3] CherryPy MemoryError

Posted: October 16th, 2010, 10:04 am
by jocke
shypike wrote:Sorry, I lost track.
No problem (-:
Does this issue result in any loss of functionality? Does the UI recover when you just refresh the browser page?
No data-loss or corruption occurs. I have to restart SABnzbd before it goes away.

Re: [0.5.3] CherryPy MemoryError

Posted: October 16th, 2010, 10:14 am
by shypike
How weird, this wasn't reported before and I never saw it myself.
Are there any specific circumstances that lead to this?
Multilpe skins, multiple computers accessing the UI?

Re: [0.5.3] CherryPy MemoryError

Posted: October 17th, 2010, 6:05 pm
by jocke
shypike wrote:Are there any specific circumstances that lead to this?
Not that I can think of. I haven't been able to reproduce the issue, and it's only happened two times so far.
Multiple skins?
Just the default ones.
Multiple computers accessing the UI?
Mainly just one, but some times there are 2 or 3 computers that access the UI at the same time. I'm not sure how many accessed the UI when the error occurred, though.

Re: [0.5.3] CherryPy MemoryError

Posted: October 18th, 2010, 2:23 am
by shypike
CherryPy is capable of handling multiple access.
I don't know how good that works, because it's very hard to test this seriously.
Looking at the code I'd say it could even be a problemĀ  in the Python runtime library.