[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Chicken-users] a strange observation
From: |
Jörg F. Wittenberger |
Subject: |
[Chicken-users] a strange observation |
Date: |
Sun, 01 Dec 2013 23:12:54 +0100 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.1.0 |
Hi all,
recently I'm stuck in a strange situation. At the moment I can't make
head or tail of it. Any hint what I could try very much appreciated!!!
The net result: some machines will work for some time and then
not-quite-freeze; they will still work somewhat as below.
Has anybody ever seen something like this? (I suppose that what I see
here is ultimately a Linux bug, after all no unix-alike system should
ever hang like this; though I'm posting it here since so far I'm afraid
I have only ever seen such a hang here and the only suspect program was
written in chicken.)
My observations:
The very same code runs on 9 machines. (3 amd64, 6 ARM) Two hang.
Always the same two.
Both Tonido plug's, that is ARM on Linux, Debian wheezy.
One of them is new, only a few weeks old. The other one served this
process (though running older code obviously) for more that a year with
no problem.
It appears that the problem began not earlier that my last update to
chicken 4.8.3 at Nov 6th.
"top" reports a load of 6.0; close to no processor load;
The only one process running a chicken built executable has about 70MB
memory; this process is not responsive at all (it's supposed serve
network requests, connection attempts just hang forever) and CANNOT be
killed even by "kill -9";
Worse: a "netstat -l -t -p" happens to hang, I can not kill that netstat
either.
So far the only thing I can do seems to be "reboot -f" (to avoid
"shutdown" getting stuck in the inability to kill that process either).
Thanks for any hints!
/Jörg
- [Chicken-users] a strange observation,
Jörg F. Wittenberger <=