Hi
We have some serious problems with alpha_dux40 file-servers running
datan# rxdebug -version localhost 7000
AFS version: Base configuration afs3.5 3.17;Izumach-12054-afs3.5-make.FileLog
[1]
The symptoms are that the file-server starts to eat out up the CPU in a
rather concerning way (ie around 99%).
When tracing the file-server we get the following:
gettimeofday (0x1404159c0, 0x0) = 0
recvmsg (5, 0x140415a30, 0) = 65
sendmsg (5, 0x140415850, 0) = 66
gettimeofday (0x140415930, 0x0) = 0
gettimeofday (0x1404159c0, 0x0) = 0
recvmsg (5, 0x140415a30, 0) = 65
sendmsg (5, 0x140415850, 0) = 66
gettimeofday (0x140415930, 0x0) = 0
gettimeofday (0x1404159c0, 0x0) = 0
recvmsg (5, 0x140415a30, 0) = 65
When tcpdumping it seams that the file-server get most
RX_PACKET_TYPE_{DATA,ACK} packets.
The output of rxdebug of rather lengthy too [2] when the server hangs.
Have anyone else seen this problem ? Its is of MAJOR problem and we
think about down-grading to 3.4 again.
Love Hornquist-Astrand
[EMAIL PROTECTED]
[1] Shouldn't it be \0 terminated somewhere after 3.17, and why didn't
the version got bumped for fix for that race condition fix ?)
[2]
Trying 130.237.48.110 (port 7000):
Free packets: 540, packet reclaims: 0, calls: 120480, used FDs: 64
not waiting for packets.
42 calls waiting for a thread
2 threads are idle
rx stats: free packets 540, allocs 567320, alloc-failures(rcv 0/0,send 0/0,ack 0)
greedy 0, bogusReads 0 (last from host 0), noPackets 0, noBuffers 0, selects 0,
sendSelects 0
packets read: data 122363 ack 168365 busy 0 abort 36 ackall 10 challenge 3 response
177 debug 116 params 0 unused 0 unused 0 unused 0 version 0
other read counters: data 122360, ack 168296, dup 15 spurious 1 dally 69
packets sent: data 200276 ack 1999 busy 0 abort 283 ackall 23 challenge 177
response 3 debug 0 params 0 unused 0 unused 0 unused 0 version 0
other send counters: ack 1999, data 884768 (not resends), resends 131, pushed 0,
acked&ignored 418114
(these should be small) sendFailed 0, fatalErrors 0
Average rtt is 0.037, with 167419 samples
Minimum rtt is 0.000, maximum is 7.272
318 server connections, 237 client connections, 237 peer structs, 485 call structs,
426 free call structs
Connection from host 130.237.48.111, port 7001, Cuid b8350043/e0f6600c
serial 5, natMTU 1444, security index 0, server conn
call 0: # 1, state precall, mode: eof, flags: waiting_for_process receive_done,
has_input_packets
call 1: # 0, state not initialized
call 2: # 0, state not initialized
call 3: # 0, state not initialized
Connection from host 130.237.48.13, port 7001, Cuid a570aa6d/df8dbc30
serial 12, natMTU 1444, security index 0, server conn
call 0: # 8, state precall, mode: eof, flags: waiting_for_process receive_done,
has_input_packets
call 1: # 0, state not initialized
call 2: # 0, state not initialized
call 3: # 0, state not initialized
[yet other 43 of these]
