tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Complete lock-up from using pkgsrc/net/darkstat



>>> [...] in case anyone can imagine how and why a complete system
>>> lockup could happen as the result of an interface being used in
>>> promiscuous mode for long periods of time (and not when used that
>>> way for short periods of time.
>> Don't forget, it may _not_ be "as the result of an interface being
>> used in promiscuous mode for long periods of time".  That's merely a
>> correlate [...]
> You're absolutely right - I don't know this for sure, but I can add
> some additional information.

> [...]

Okay, so darkstat is a pretty strong correlate; I'd be surprised if the
hangs didn't have _something_ to do with it.  But the problem may be
related to the network only incidentally (as in, it's possible the
lockups have nothing to do with the network beyond the program that
provokes them happening to run the network promiscuous).  For example,
maybe darkstat manages its memory in an unusual way that tickles a bug
in your port's pmap layer - that's an unlikely scenario, and it's not
too unlikely the issue *is* network-related; I'm just warning against
writing off possibilities without evidence.  Too often I've spent days
scratching my head trying to find a problem in thing A only to later
discover the actual problem is over in thing B instead.

>> It also might be relevant to note which port you're running.  It
>> must be capable of having re and wm interfaces, since you name them,
>> but that still includes a fair bit.
> With five different machines, I don't think it's likely an issue with
> all five interfaces, but it's always better to have too much
> information than too little:

Yes.  (Well, at least up to the point where extra information makes it
difficult to find useful information, I would say but the list posts
are nowhere near that point yet.)

I note they're all wm or re.  If it *does* have something to do with
the network, it could be something interface-independent, or it could
be something related to a small subset of network drivers, a set that
includes wm and re.

I still see no statement which port(s) you're running, except that your
dmesg quotes indicate some machines have msix interrupt handling and
some (others?) have ioapic.  (Of course, it's entirely possible the
issue has nothing to do with which port you're using.  Again, I'm just
hesitant to write off possible causes without evidence.)

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse%rodents-montreal.org@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B


Home | Main Index | Thread Index | Old Index