Le 17/11/2011 14:30, Win Htin a écrit :
> Hi,
>
> I haven't tried on LTO5 but on LTO4, I need to run the following "mt"
> parameters for the tape-drives to function properly.
> Hope this helps.
>
> Below, replace<tapedevice> with<st1> in your scenario.
>
> ----------------------------------- Start
> --------------------------------------------------------
> 1. mt -f /dev/<tapedevice> stsetoptions
>
> 2. mt -f /dev/<tapedevice> compression
>
> 3. mt -f /dev/<tapedevice> stsetoptions buffer-writes
>
> 4. mt -f /dev/<tapedevice> stsetoptions async-write
>
> 5. mt -f /dev/<tapedevice> stsetoptions fast-eod
>
> 6. mt -f /dev/<tapedevice> stsetoptions two-fms
>
> 7. tapeinfo -f /dev/<autochanger> <=== (/dev/sg3 in your scenario)
>
> ## Following is the output from my library from the above command.
> Product Type: Tape Drive
> Vendor ID: 'IBM '
> Product ID: 'ULT3580-TD4 '
> Revision: 'xxxxxx'
> Attached Changer: No
> SerialNumber: 'xxxxxxxxx'
> MinBlock:1
> MaxBlock:16777215
> SCSI ID: 2
> SCSI LUN: 0
> Ready: yes
> BufferedMode: yes
> Medium Type: 0x48
> Density Code: 0x46
> BlockSize: 0
> DataCompEnabled: yes
> DataCompCapable: yes
> DataDeCompEnabled: yes
> CompType: 0x1
> DeCompType: 0x1
> Block Position: 7920778
> ----------------------------------------------------------- End
> -----------------------------------------------------------
#### Autochanger #####
tapeinfo -f /dev/sg3
Product Type: Medium Changer
Vendor ID: 'QUANTUM '
Product ID: 'Scalar i40-i80 '
Revision: '135G'
Attached Changer API: No
SerialNumber: 'xxxxx '
SCSI ID: 0
SCSI LUN: 1
Ready: yes
#### Drive 1 ####
tapeinfo -f /dev/sg2
Product Type: Tape Drive
Vendor ID: 'HP '
Product ID: 'Ultrium 5-SCSI '
Revision: 'Z58Z'
Attached Changer API: No
SerialNumber: 'xxxx'
TapeAlert[32]: Interface: Problem with SCSI interface between tape
drive and initiator.
MinBlock: 1
MaxBlock: 16777215
SCSI ID: 0
SCSI LUN: 0
Ready: yes
BufferedMode: yes
Medium Type: Not Loaded
Density Code: 0x58
BlockSize: 0
DataCompEnabled: yes
DataCompCapable: yes
DataDeCompEnabled: yes
CompType: 0x1
DeCompType: 0x1
BOP: yes
Block Position: 0
Partition 0 Remaining Kbytes: 1459056
Partition 0 Size in Kbytes: 1459056
ActivePartition: 0
EarlyWarningSize: 0
NumPartitions: 0
MaxPartitions: 1
### drive 2 #####
root@backup1:~# tapeinfo -f /dev/sg4
Product Type: Tape Drive
Vendor ID: 'HP '
Product ID: 'Ultrium 5-SCSI '
Revision: 'Z58Z'
Attached Changer API: No
SerialNumber: 'xxxx'
MinBlock: 1
MaxBlock: 16777215
SCSI ID: 0
SCSI LUN: 0
Ready: no
>
> Cheers,
> Win
>
>
> ------------------------------
> Message: 20
> Date: Thu, 17 Nov 2011 07:44:35 -0500
> From: Dan Langille<dan AT langille DOT org>
> Subject: Re: [Bacula-users] Fatal error: Unable to write EOF
> To: Ghislain Escorne<Ghislain.Escorne AT ujf-grenoble DOT fr>
> Cc: Bacula-users AT lists.sourceforge DOT net
> Message-ID:<9B56C61B-080F-41BB-AC7D-A30AFC093FAD AT langille DOT org>
> Content-Type: text/plain; charset="windows-1252"
>
> On Nov 17, 2011, at 4:46 AM, Ghislain Escorne wrote:
>
>> Hello !
>>
>> I have bacula Version: 5.2.1.
>> I want to backup datas on a autoloader quantum i80 with 2 drives.
> What OS are you using? That can affect the configuration.
>
>> The bacula test works correctly:
>> root@backup1:~# btape -c bacula-sd.conf /dev/nst0
>> .
>> btape: btape.c:476 open device "Drive-i80-1" (/dev/nst0): OK
>> *test
>> ...
>> We should be in file 5. I am at file 5. This is correct!
>> Do you wish to continue with the Autochanger test? (y/n): y
>> === Autochanger test ===
>> The test autochanger worked!!
>>
>> But when I run a job I have this mess :
>>
>> 15-Nov 23:33 backup1-sd JobId 3: Writing spooled data to Volume. Despooling
>> 107,374,192,579 bytes ?
> Do small local jobs work? When setting up a new install, I prefer to
> start small and increase job size later.
>
> By small local, I mean back up the machine upon which the SD is
> located. To keep things simple, do not use spooling for that job.
>
>> 15-Nov 23:56 backup1-sd JobId 3: Fatal error: Unable to write EOF.
>> ERR=dev.c:1551 ioctl MTWEOF error on "Drive-i80-1" (/dev/nst0).
>> ERR=Input/output error.
> There's lots in Google about this error, but I stopped looking when I
> didn't locate a suggestion right away.
>
> There were suggestions to activate a sleep in the mtx-changer file.
> However, it seems that writing has already occurred by the time this
> error has occurred. The sleep command is usually added to let the tape
> mount full complete before writing starts.
>
>> 15-Nov 23:56 backup1-sd JobId 3: Fatal error: spool.c:302 Fatal append error
>> on device "Drive-i80-1" (/dev/nst0): ERR=dev.c:1551 ioctl MTWEOF error on
>> "Drive-i80-1" (/dev/nst0). ERR=Input/output error.
>>
>> 15-Nov 23:56 backup1-sd JobId 3: Despooling elapsed time = 00:21:43,
>> Transfer rate = 82.40 M Bytes/second
>> 15-Nov 23:56 backup1-sd JobId 3: Job write elapsed time = 06:49:12, Transfer
>> rate = 17.47 M Bytes/second
>> 15-Nov 23:56 backup1-fd JobId 3: Error: bsock.c:389 Write error sending
>> 65536 bytes to Storage daemon:backup1:9103: ERR=Connection reset by peer
>> 15-Nov 23:56 backup1-fd JobId 3: Fatal error: backup.c:1144 Network send
>> error to SD. ERR=Connection reset by peer
> Usually, these errors indicate a network problem?. but with the error
> above, I think not.
>
>> 15-Nov 23:56 backup1-dir JobId 3: Error: Bacula backup1-dir 5.2.1 (30Oct11):
>>
>> This is my conf :
>> Autochanger {
>> Name = i80
>> Device = Drive-i80-1, Drive-i80-2
>> #Changer Command = "/home/kern/bacula/bin/mtx-changer %c %o %S %a %d"
>> Changer Command = "/etc/bacula/mtx-changer %c %o %S %a %d"
>> # %c = changer device
>> # %o = command (unload|load|loaded|list|slots)
>> # %S = slot index (1-based)
>> # %a = archive device (i.e., /dev/sd* name for tape
>> drive)
>> # %d = drive index (0-based)
>> Changer Device = /dev/sg3
>> }
>>
>> Device {
>> Name = Drive-i80-1
>> Drive Index = 0
>> ##Device Type = Tape
>> Media Type = LTO5
>> Archive Device = /dev/nst0
>> Autochanger = yes
>> LabelMedia = no;
>> AutomaticMount = yes;
>> AlwaysOpen = yes;
>> # RemovableMedia = yes;
>> # RandomAccess = no;
>> Spool Directory = /data/bacula/spool
> After you get writes working, consider using different directories for
> each drive. I can't recall why or even if this is a good suggestion.
>> Maximum Spool Size = 100 G #
>> # Autoselect = yes
>> Maximum Changer Wait = 6000
>> Alert Command = "sh -c 'tapeinfo -f %c |grep TapeAlert|cat'"
>> }
>>
>>
>> Device {
>> Name = Drive-i80-2
>> Drive Index = 1
>> ## Device Type = Tape
>> Media Type = LTO5
>> Archive Device = /dev/nst1
>> Autochanger = yes
>> LabelMedia = no;
>> AutomaticMount = yes;
>> AlwaysOpen = yes;
>> # RemovableMedia = yes;
>> # RandomAccess = no;
>> Spool Directory = /data/bacula/spool
>> Maximum Spool Size = 100 G #
>> # Autoselect = yes
>> Maximum Changer Wait = 6000
>> Alert Command = "sh -c 'tapeinfo -f %c |grep TapeAlert|cat'"
>> }
>>
>> Thank you for your help
>>
>> ------------------------------------------------------------------------------
>> All the data continuously generated in your IT infrastructure
>> contains a definitive record of customers, application performance,
>> security threats, fraudulent activity, and more. Splunk takes this
>> data and makes sense of it. IT sense. And common sense.
>> http://p.sf.net/sfu/splunk-novd2d_______________________________________________
>> Bacula-users mailing list
>> Bacula-users AT lists.sourceforge DOT net
>> https://lists.sourceforge.net/lists/listinfo/bacula-users
------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users
|
