[Samba] strange situation

[Jason Greene]

Which log do you want output from? The samba logs didn't change when I ran
that.

here are the results

[root at s06b-par01l ~]# netstat -s
Ip:
    82262 total packets received
    0 forwarded
    0 incoming packets discarded
    58038 incoming packets delivered
    48287 requests sent out
Icmp:
    161 ICMP messages received
    1 input ICMP message failed.
    ICMP input histogram:
        destination unreachable: 7
        echo requests: 143
        echo replies: 11
    210 ICMP messages sent
    0 ICMP messages failed
    ICMP output histogram:
        destination unreachable: 67
        echo replies: 143
Tcp:
    430 active connections openings
    173 passive connection openings
    1 failed connection attempts
    48 connection resets received
    20 connections established
    53189 segments received
    45120 segments send out
    41 segments retransmited
    0 bad segments received.
    455 resets sent
Udp:
    3569 packets received
    65 packets to unknown port received.
    0 packet receive errors
    2946 packets sent
TcpExt:
    5 resets received for embryonic SYN_RECV sockets
    7 packets pruned from receive queue because of socket buffer overrun
    ArpFilter: 0
    226 TCP sockets finished time wait in fast timer
    670 delayed acks sent
    Quick ack mode was activated 991 times
    109 packets directly queued to recvmsg prequeue.
    32269 packets directly received from prequeue
    20242 packets header predicted
    41 packets header predicted and directly queued to user
    TCPPureAcks: 3539
    TCPHPAcks: 7867
    TCPRenoRecovery: 0
    TCPSackRecovery: 0
    TCPSACKReneging: 0
    TCPFACKReorder: 0
    TCPSACKReorder: 0
    TCPRenoReorder: 0
    TCPTSReorder: 0
    TCPFullUndo: 0
    TCPPartialUndo: 0
    TCPDSACKUndo: 0
    TCPLossUndo: 4
    TCPLoss: 0
    TCPLostRetransmit: 0
    TCPRenoFailures: 0
    TCPSackFailures: 0
    TCPLossFailures: 0
    TCPFastRetrans: 0
    TCPForwardRetrans: 0
    TCPSlowStartRetrans: 0
    TCPTimeouts: 13
    TCPRenoRecoveryFail: 0
    TCPSackRecoveryFail: 0
    TCPSchedulerFailed: 0
    TCPRcvCollapsed: 245
    TCPDSACKOldSent: 1115
    TCPDSACKOfoSent: 824
    TCPDSACKRecv: 0
    TCPDSACKOfoRecv: 0
    TCPAbortOnSyn: 0
    TCPAbortOnData: 1
    TCPAbortOnClose: 43
    TCPAbortOnMemory: 0
    TCPAbortOnTimeout: 4
    TCPAbortOnLinger: 0
    TCPAbortFailed: 0
    TCPMemoryPressures: 0

On Thu, May 29, 2008 at 12:56 PM, Scott Lovenberg <scott.lovenberg at gmail.com>
wrote:

> Jason Greene wrote:
>
>> smbd version 3.0.25b-0.4E.5
>>
>> Our server was functioning very well for several months.  Our SAN crapped
>> out and the LUN the server was using was gone.  Everything is back up
>> except
>> SAMBA is acting crazy.
>>
>> I am looking at the logs and I am getting
>>
>> /var/log/samba/winbindd.log
>> winbindd: Exceeding 200 client connections, no idle connection found
>> and
>> ads_krb5_mk_req: krb5_get_credentials failed for
>> e05r5s-dc02$@ENT.DS.GSA.GOV(Server not found in Kerberos database)
>>
>>
>>
>> this on is strange because I get this in the log
>> /var/log/samba/wb-ENT.log
>> error getting user info for sid
>> S-1-5-21-1482476501-413027322-682003330-143384
>>
>> but when I do this
>>
>> wbinfo -s S-1-5-21-1482476501-413027322-682003330-143394
>> ENT+(User Name)  (edited out user name)
>>
>>
>>  /var/log/samba/winbindd-dc-connect.log
>> [2008/05/29 12:12:11, 1] libsmb/clientgen.c:cli_rpc_pipe_close(387)
>>  cli_rpc_pipe_close: cli_close failed on pipe \NETLOGON, fnum 0x800b to
>> machine s06b-fin-dc02.finance.int.  Error was SUCCESS - 0
>>
>> /var/log/samba/wb-FINANCE.log <==
>> [2008/05/29 12:25:27, 0] lib/util_tdb.c:tdb_log(662)
>>  tdb(/var/cache/samba/netsamlogon_cache.tdb): tdb_rec_read bad magic
>> 0xd9fee666 at offset=27920
>>
>>
>> If I look at ps ax I get about 20 winbind entries
>>
>> When we try to access the only share on the box everything seems to
>> hang...
>> we can't even do and ls -al
>> Then I restart winbind and everything frees up.
>>
>> It seems as if winbind is hanging
>>
>> Anyone know whet might be going on and how to resolve it?
>>
>>
>>
>>
>>
>>
>>
> Anything interesting in 'netstat -s'?  This sounds a bit like something I
> was seeing with a corrupted E1000 nic module...  It was timing out just
> about every connection.
> Could you post a section of your logs output while this is happening?
>



-- 
Jason