Closed
Bug 164716
Opened 22 years ago
Closed 6 years ago
[mozbot.pl] mozbot killed by flooding
Categories
(Webtools Graveyard :: Mozbot, defect)
Webtools Graveyard
Mozbot
Tracking
(Not tracked)
RESOLVED
INACTIVE
People
(Reporter: bbaetz, Unassigned)
Details
Attachments
(1 file)
1.32 KB,
text/plain
|
Details |
After a few floods, it just dies, with './mozbot.pl: killed' at the prompt
Theres nothing in the logs, so maybe this is a Net::IRC bug (ssdbot is running
NET::IRC 0.73)
I tried enabling core dumps via ulimit (well, the csh version of that), but I
still couldn't get a core file to be produced.
landifll had a forked process doing tinderbox updates all teh time; not sure if
that was a fallout from that or the other bug (it was an older one)
This is not the same as the 'ping timeout' bug, because there mozbot keeps running.
justdave can probably let you have access to the logs on landfill, but the tail
of the log (pid 21559 for later reference) constists of about 40 SpottedJoins in
5 seconds, then a couple of Heards, then 5 CTCPPing in two seconds, then nothing.
This has happened a couple of times in the past few days; I'll try (again)
setting ssdbot up so that it doesn't join to #mozilla, and then we probably
won't see any of this. Which is just working arround hte issue, but....
Comment 1•22 years ago
|
||
I would guess it's hitting a CPU or memory limit. I'll keep an eye out for it,
but I'll need a server where flooding is common to test this... any ideas?
Updated•22 years ago
|
Summary: mozbot killed by flooding → [mozbot.pl] mozbot killed by flooding
I dunno if this is realted to this bug or not, but on the mozbot I run, I had a few
'l'users that kept making it crash by doing a ctcp ping 99999999999 to the bot. I
use using the mozbot-latest sources.
(In reply to comment #2)
> I dunno if this is realted to this bug or not, but on the mozbot I run, I had a
> few
> 'l'users that kept making it crash by doing a ctcp ping 99999999999 to the bot.
> I
> use using the mozbot-latest sources.
>
It is. It isnt the $ctime value that causes it it is the fact they are flooding it. If I ping it 5 times or so very quickly it forkbombs so badly I almost have to restart the machine.
Also, I managed to get this from mozbot while trying to flood it.
Use of uninitialized value in string eq at
/usr/lib/perl5/site_perl/5.8/Net/IRC/Connection.pm line 1588 (#1)
I got it by flooding it with /msg lol! so many times it would say long send queue and then pinging quickly 3 times. While I was typing this it did the following:
2006-01-22 10:08:10 UTC (1692) eek! disconnected from network: 'ERROR :Closing Link: testbot[192.168.1.2] (Ping timeout)'
(reconnecting garbage here)
After reconnecting successfully it sat there about 2 minutes and then:
"Segmentation fault (core dumped)"
I dont know if its useful but I will attach it.
Updated•19 years ago
|
QA Contact: kerz → mozbot
Updated•16 years ago
|
Severity: critical → major
OS: Linux → All
Hardware: x86 → All
Mozbot/Firebot is in maintenance-only mode; no further changes are planned.
Status: NEW → RESOLVED
Closed: 6 years ago
Resolution: --- → INACTIVE
Updated•6 years ago
|
Product: Webtools → Webtools Graveyard
You need to log in
before you can comment on or make changes to this bug.
Description
•