[pgsql-jp: 37090] AccessからのODBC接続が不安定

[Hisashi Chiba]

千葉と申します。

現在、RedHat7.3、PostgreSQL-7.3.2の環境でWindowsクライアントからpsqlodbc
-8.01.200 ODBCドライバ経由で接続しています。

Accessアプリケーションではリンク接続で数日前まで問題なく参照出来て
いたのですが、先ほどODBCエラーが発生して接続できない状態が発生し、
現在は接続自体は問題ないようです。

心配になってサーバーのシステムログを見てみると、下のような気がかり
なメッセージが残っていました。

何となく、HDが壊れたようにも思えるのですが、postmasterが自動で
リカバリーされて再起動することもあるのでしょうか？

識者の方々の考えをお聞かせ頂ければ、有り難く存じます

# less /var/log/messages
----------（抜粋）--------
.
.
May 22 22:48:55 server001 kernel: scsi : aborting command due to timeout
: pid 28196644, scsi0, channel 0, id 0, lun 0 Write (10) 0
0 02 1e 27 2a 00 00 d0 00
May 22 22:48:55 server001 postgres[11999]: [27] PANIC:  open of
/usr/local/pgsql/data/pg_clog/0023 failed: ??????????????????????
May 22 22:48:55 server001 kernel: scsi : aborting command due to timeout
: pid 28196646, scsi0, channel 0, id 0, lun 0 Write (10) 0
0 02 1e 27 fa 00 00 d0 00
May 22 22:48:55 server001 postgres[11999]: [28] LOG:  statement: analyze;
May 22 22:48:55 server001 kernel: scsi : aborting command due to timeout
: pid 28196649, scsi0, channel 0, id 0, lun 0 Write (10) 0
0 02 1e 28 ca 00 00 30 00
May 22 22:48:55 server001 postgres[1155]: [25] LOG:  server process (pid
11999) was terminated by signal 6
May 22 22:48:55 server001 postgres[1155]: [25] LOG:  server process (pid
11999) was terminated by signal 6
May 22 22:48:56 server001 postgres[1155]: [26] LOG:  terminating any
other active server processes
May 22 22:48:56 server001 postgres[12004]: [27] LOG:  connection
received: host=[local]
May 22 22:48:56 server001 postgres[11957]: [27-1] WARNING:  Message from
PostgreSQL backend:
May 22 22:48:56 server001 postgres[12004]: [28] FATAL:  The database
system is in recovery mode
May 22 22:48:56 server001 postgres[11957]: [27-2] ^IThe Postmaster has
informed me that some other backend
May 22 22:48:56 server001 postgres[12006]: [27] LOG:  connection
received: host=[local]
May 22 22:48:56 server001 postgres[11957]: [27-3] ^Idied abnormally and
possibly corrupted shared memory.
May 22 22:48:56 server001 postgres[12006]: [28] FATAL:  The database
system is in recovery mode
May 22 22:48:56 server001 postgres[11957]: [27-4] ^II have rolled back
the current transaction and am
May 22 22:48:56 server001 postgres[11957]: [27-5] ^Igoing to terminate
your database system connection and exit.
May 22 22:48:56 server001 postgres[11957]: [27-6] ^IPlease reconnect to
the database system and repeat your query.
May 22 22:48:56 server001 postgres[1155]: [27] LOG:  all server
processes terminated; reinitializing shared memory and semaphores
.
.
May 24 09:45:41 server001 postgres[16535]: [55] LOG:  connection
received: host=192.168.1.105 port=1280
May 24 09:45:41 server001 postgres[16535]: [56] FATAL:  unsupported
frontend protocol
May 24 09:45:41 server001 postgres[16536]: [55] LOG:  connection
received: host=192.168.1.105 port=1281
May 24 09:45:41 server001 postgres[16536]: [56] LOG:  connection
authorized: user=Admin database=eigyo
May 24 09:45:41 server001 postgres[16536]: [57] FATAL:  user "Admin"
does not exist
May 24 09:45:41 server001 postgres[16537]: [55] LOG:  connection
received: host=192.168.1.105 port=1282
May 24 09:45:41 server001 postgres[16537]: [56] FATAL:  unsupported
frontend protocol
May 24 09:45:41 server001 postgres[16538]: [55] LOG:  connection
received: host=192.168.1.105 port=1283
May 24 09:45:41 server001 postgres[16538]: [56] LOG:  connection
authorized: user=postgres database=eigyo
May 24 09:45:41 server001 postgres[16538]: [57] ERROR:  Relation
"msysconf" does not exist
May 24 09:45:41 server001 postgres[1155]: [55] LOG:  server process (pid
16538) was terminated by signal 11
May 24 09:45:41 server001 postgres[1155]: [56] LOG:  terminating any
other active server processes
May 24 09:45:41 server001 postgres[1155]: [57] LOG:  all server
processes terminated; reinitializing shared memory and semaphores
May 24 09:45:41 server001 postgres[16539]: [58] LOG:  database system
was interrupted at 2006-05-24 09:44:54 JST
May 24 09:45:41 server001 postgres[16539]: [59] LOG:  checkpoint record
is at 23/C2751E84
May 24 09:45:41 server001 postgres[16539]: [60] LOG:  redo record is at
23/C2751E84; undo record is at 0/0; shutdown TRUE
May 24 09:45:41 server001 postgres[16539]: [61] LOG:  next transaction
id: 3898497; next oid: 58839596
May 24 09:45:41 server001 postgres[16539]: [62] LOG:  database system
was not properly shut down; automatic recovery in progress
May 24 09:45:41 server001 postgres[16539]: [63] LOG:  ReadRecord: record
with zero length at 23/C2751EC4
May 24 09:45:41 server001 postgres[16539]: [64] LOG:  redo is not required
May 24 09:45:43 server001 postgres[16539]: [65] LOG:  database system is
ready
-------------------------------