connection issues with squid2-head/squid3-head

Hi,

I've been polygraphing squid3 over the last few days and I've
found a few niggling bugs, most of which I've been able to fix.

However, there's quite a big one which seems to screw up polygraph:

000.01| group-id: 07583985.4a77160d:00000002 pid: 5645
000.01| current time: 1035568083.700321 or Fri, 25 Oct 2002 17:48:03 GMT
000.01| fyi: current state (1) stored
000.01| starting 1 HTTP agents...
000.01| starting agent[1 / 07583985.4a77160d:00000004] `DataComm-1-rbt' on 10.61.2.8
000.04| ./Connection.cc:175: error: 1/1 (54) Connection reset by peer
000.04| error reading from 10.61.2.20:3128
000.05| error reading from 10.61.2.20:3128
000.06| ./Connection.cc:175: error: 3/3 (54) Connection reset by peer
000.06| error reading from 10.61.2.20:3128
000.07| ./Connection.cc:175: error: 7/7 (54) Connection reset by peer
000.07| error reading from 10.61.2.20:3128
000.07| ./Connection.cc:175: error: 15/15 (54) Connection reset by peer
000.07| error reading from 10.61.2.20:3128
000.08| fyi: server scan completed with all 1 local robots ready to hit all 1 servers
000.08| fyi: local phase `wait' reached synchronization point
000.09| p-wait 302 76.87 817 58.94 0
000.09| ./Connection.cc:175: error: 31/31 (54) Connection reset by peer
000.09| error reading from 10.61.2.20:3128
000.10| i-warm 331 77.38 906 56.19 0 202
000.14| ./Connection.cc:175: error: 63/63 (54) Connection reset by peer
000.14| error reading from 10.61.2.20:3128
000.18| i-warm 811 110.38 2214 24.38 0 202
000.20| ./Connection.cc:175: error: 127/127 (54) Connection reset by peer
000.20| error reading from 10.61.2.20:3128
000.20| fyi: new oid storage is primed
000.26| i-warm 1352 122.00 1947 37.34 0 185
000.35| i-warm 1864 114.79 1715 45.70 0 168
000.39| ./Connection.cc:175: error: 255/255 (54) Connection reset by peer
000.39| error reading from 10.61.2.20:3128
000.43| i-warm 2400 120.57 1520 49.63 0 145
000.51| i-warm 2897 109.39 1468 53.32 0 155

.. at ~100req/sec. Now, the HEAD tag beforeClientStreams behaves
fine, but postClientStreams and anything since is showing the above
problems with polygraph-2.5.7.

Its just a heads up - hopefully Robert and I can sort this out over the
weekend. I'll take a look at the changes I've placed into squid3 and
see if they're needed in squid2-head/squid-2-5 (I think at least one
patch will be needed in all..)

Have fun, and long live evil C++.

Adrian
Received on Fri Oct 25 2002 - 02:06:48 MDT