tor eating CPU time [signed]

[Michael Laccetti [c]]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Well, I don't have much in terms of profiling, but I have a nice overview of
what's happening.

Side note: most of the SSL certs I use are signed by cacert.org, so shoot on
over there and grab their root cert if you haven't already.  And tell your
friends.  :)

Nirvana traffic:
https://nirvana.s2g-limited.com/~nihilist/nirvana_traffic.png
You'll see where I restarted the tor service after it had last crashed.

Nirvana CPU:
https://nirvana.s2g-limited.com/~nihilist/nirvana_cpu.png 
CPU corresponded to when I fired up 0.9.3.

Nirvana load:
https://nirvana.s2g-limited.com/~nihilist/nirvana_load.png
Again, big upward spike.

Actually had to take it offline, it was impacting the rest of the stuff that
was on the server.  If 1.0.0-alpha-cvs stabilizes sometime soon, let me know,
and I'll fire it up again.  Or, if it'll take more then a week or so, when the
rest off the stuff is ported off the server, it can be a dedicated tor node
again.

Hope the info helps.

- -----Original Message-----
From: owner-or-talk at freehaven.net [mailto:owner-or-talk at freehaven.net] On
Behalf Of Roger Dingledine
Sent: January 28, 2005 07:16
To: or-talk at freehaven.net
Subject: Re: tor eating CPU time [u]

On Fri, Jan 28, 2005 at 12:56:12PM +0100, Valient Gough wrote:
> I upgraded to 0.0.9.3 last weekend, and it dropped tor's average CPU 
> usage by almost half.

Great.

> However thing have been slowly getting worse.  Saturday stats show 
> about 30% CPU average usage, increasing slowly to a little over 40% on 
> Wed, then about 50% by Thursday, and about 12 hours ago tor jumped to 
> 100% cpu usage and has not taken a rest since then.
> 
> The strange thing is that this doesn't correlate with traffic served.  
> Tor traffic has been steadily averaging about 200k/sec, and started 
> dropping about 12 hours ago by as much as a factor of 2 (much higher 
> variation though).  So, CPU time increased, but network traffic decreased...
> 
> Looking through my server monitoring logs, I don't see any correlation 
> against network usage or number of open TCP connections.   I'm about to 
> restart TOR, but before I get into the habbit of periodic restarts, is 
> there anything I can do to get information about the problem?

As near as I can tell it, Tor spends most of its time doing SSL handshakes with
other Tors (either clients or servers).

I don't think anybody has given me profiling output lately (gprof, oprofile,
etc) on a really busy server though. If you wanted to give that a go and let me
know where it spends its time _once it gets to 100% cpu use_, that would be
really handy.

For example, I can imagine a couple situations where it would start doing
linear lookups on a list of circuits that has grown really large -- and it
would do such a lookup for each cell that it handles. But until I know what's
going on, trying to optimize everything is just going to increase complexity.

Thanks,
- --Roger


-----BEGIN PGP SIGNATURE-----
Version: PGP 8.1

iQA/AwUBQfrtN7sNGE8hfSdaEQIRlQCdGMZ1BrylNU1/pBSyEgxLM0OOHLoAoO/k
134WPJz4tPSt/iHJAmz0ZiDo
=NTYm
-----END PGP SIGNATURE-----


-- 
---------------------[ Ciphire Signature ]----------------------
From: michael at s2g-limited.com signed email body (2587 characters)
Date: on 29 January 2005 at 01:56:07 UTC
To:   or-talk at freehaven.net
----------------------------------------------------------------
: Ciphire has secured this email against identity theft.
: Free download at www.ciphire.com. The garbled lines
: below are the sender's verifiable digital signature.
----------------------------------------------------------------
00fAAAAAEAAAA37fpBGwoAAAEDAAIAAgACACCF2BCZF4VOp3SsEUZceUhJFcF28h
RH2mHTTNijCotTgAEAemslQ+vzEgqJDuYPs8gA9aPwJyt5Y1+3STnOLkRy0U1LGM
stNnj4moBkEw6aXPqJ6a7MWLaEzgw18fOlUsCnxg==
------------------[ End Ciphire Signed Message ]----------------