[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Pan-users] Odd issue with latest git code
From: |
walt |
Subject: |
Re: [Pan-users] Odd issue with latest git code |
Date: |
Thu, 10 Apr 2014 15:26:15 -0700 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.4.0 |
On 04/10/2014 09:56 AM, Jim Henderson wrote:
> It may be present in earlier code as well.
>
> If pan sits idle for a while, at least on openSUSE 13.1 running GNOME3
> (gtk2 build of pan, though), it seems to eventually become non-
> responsive. It seems to be related to the amount of time it's running -
> I've (for example) started replying to a post, left the machine and come
> back, and the only option is to force close pan and start over (losing my
> previous partial post content).
>
> I left it running last night, and had to force close it again this
> morning before I could do anything.
>
> Anyone else seeing this?
I'm seeing sudden halts while downloading headers or articles, but I don't
normally let pan sit idle so I can't address your specific use-case.
I suggest you run pan from gdb (from a bash shell prompt) and look for
error messages.
I see one of two errors when I run pan from gdb: either a corrupted double-
linked list (hard to imagine if pan is sitting idle) or a segfault, usually
involving gnutls.
Come to think of it, the segfault causes pan to close, but the corrupt linked
list error just causes pan to be unresponsive while still running.
I have trouble seeing how an idle pan could cause a linked-list corruption,
but my imagination isn't what it used to be :/