Veritas-bu

[Veritas-bu] IBM LTO-4 Encryption, Drive Reset/Errors?

2009-08-18 09:18:43
Subject: [Veritas-bu] IBM LTO-4 Encryption, Drive Reset/Errors?
From: Justin Piszcz <jpiszcz AT lucidpixels DOT com>
To: veritas-bu AT mailman.eng.auburn DOT edu
Date: Tue, 18 Aug 2009 09:15:34 -0400 (EDT)
Hello,

Generally all of our media servers use QLogic HBAs:

In one environment, we see this, with IBM LTO-4 drives w/ENCRYPTION 
enabled:

st1: Error 20000 (sugg. bt 0x0, driver bt 0x0, host bt 0x2).
qla2xxx 0000:22:00.0: LOOP DOWN detected (2).
qla2xxx 0000:22:00.0: LIP reset occured (f7f7).
qla2xxx 0000:22:00.0: LIP occured (f7f7).
qla2xxx 0000:22:00.0: LOOP UP detected (4 Gbps).

Both StorageTek, er, Sun, er, Oracle and HP both came out to diagnose the 
hardware, HP replaced the fiber cables, Sun verified the drives were OK:

bptm log output:

12:57:39.299 [1048] <2> write_data: block position check: actual 3973524, 
expected 3973524
12:57:39.299 [1048] <2> io_write_back_header: drive index 0, 
box.hostname.fqdn_1250598835, file num = 39, mpx_headers = 0, copy 1
12:57:39.302 [1048] <2> io_write_back_header: encryption status: nexus scope 1, 
key scope 1
12:57:39.302 [1048] <2> io_write_back_header: encryp mode 0x0, decryp mode 0x0, 
algorithm index 1, key instance 98
12:57:39.305 [1048] <2> write_data: completed writing backup header, start 
writing data when first buffer is available, copy 1
12:57:39.325 [1048] <2> manage_drive_encryption: encryption status: nexus scope 
1, key scope 1
12:57:39.325 [1048] <2> manage_drive_encryption: encryp mode 0x2, decryp mode 
0x0, algorithm index 1, key instance 99
12:57:39.325 [1048] <2> manage_drive_encryption: Kad type 0x0, Kad length 32 
Kad [hash]
12:57:39.325 [1048] <2> write_data: first write, twin_index: 0 cindex: 0 
dont_process: 1 wrote_backup_hdr: 1 finished_buff: 0
12:57:39.325 [1048] <2> write_data: received first buffer (262144 bytes), begin 
writing data
12:58:01.247 [1048] <2> signal_parent: sending SIGUSR1 to bpbrm (pid = 1042)
12:58:01.247 [1048] <2> write_data: attempting write error recovery, err = 5
12:58:01.247 [1048] <2> tape_error_rec: error recovery to block 3976640 
requested
12:58:01.247 [1048] <2> tape_error_rec: attempting error recovery, delay 3 
minutes before next attempt, tries left = 5
13:01:01.269 [1048] <2> io_ioctl: command (5)MTWEOF 0 from (overwrite.c.503) on 
drive index 0
13:01:52.368 [2606] <2> bptm: INITIATING (VERBOSE = 1): -rptdrv -jobid 
-1249386471 -jm 
13:01:52.382 [2606] <2> drivename_open: Called with Create 0, file Drive01
13:01:52.382 [2606] <2> drivename_checklock: Called
13:01:52.382 [2606] <2> drivename_checklock: PID 878 has lock

I am now re-running the backups with ENCRYPTION disabled as a test as I 
have never seen such behavior before, has anyone seen this issue utilizing 
the built-in encryption on IBM LTO-4 drives?

OS = RHEL5 x86_64
Fiber Card = QLogic PCI-Express Dual Port 4Gbps Card

Justin.

_______________________________________________
Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu

<Prev in Thread] Current Thread [Next in Thread>
  • [Veritas-bu] IBM LTO-4 Encryption, Drive Reset/Errors?, Justin Piszcz <=