gjournal: what is it good for?

David N davidn04 at gmail.com
Fri Apr 23 01:09:02 UTC 2010


On 21 April 2010 21:47, David N <davidn04 at gmail.com> wrote:
> On 21 April 2010 19:50, Andrew Reilly <areilly at bigpond.net.au> wrote:
>> Hi David,
>>
>> On Wed, Apr 21, 2010 at 07:16:08PM +1000, David N wrote:
>>> Can you show me a print out of
>>> gjournal list
>>
>> here:
>>
>>
>> Geom name: gjournal 827322917
>> ID: 827322917
>> Providers:
>> 1. Name: mirror/gm0e.journal
>>   Mediasize: 981949479936 (915G)
>>   Sectorsize: 512
>>   Mode: r1w1e1
>> Consumers:
>> 1. Name: mirror/gm0e
>>   Mediasize: 983023222272 (916G)
>>   Sectorsize: 512
>>   Mode: r1w1e1
>>   Jend: 983023221760
>>   Jstart: 981949479936
>>   Role: Data,Journal
>>
>> Geom name: gjournal 1843456033
>> ID: 1843456033
>> Providers:
>> 1. Name: ad8s1a.journal
>>   Mediasize: 1499228127232 (1.4T)
>>   Sectorsize: 512
>>   Mode: r1w1e1
>> Consumers:
>> 1. Name: ad8s1a
>>   Mediasize: 1500301869568 (1.4T)
>>   Sectorsize: 512
>>   Mode: r1w1e1
>>   Jend: 1500301869056
>>   Jstart: 1499228127232
>>   Role: Data,Journal
>>
>> Geom name: gjournal 3015273731
>> ID: 3015273731
>> Providers:
>> 1. Name: ad10s1.journal
>>   Mediasize: 749081051136 (698G)
>>   Sectorsize: 512
>>   Mode: r1w1e1
>> Consumers:
>> 1. Name: ad10s1
>>   Mediasize: 750154793472 (699G)
>>   Sectorsize: 512
>>   Mode: r1w1e1
>>   Jend: 750154792960
>>   Jstart: 749081051136
>>   Role: Data,Journal
>>
>>
>>> and
>>> bsdlabel /dev/mirror/gm0e.journal
>>
>> duncan [201]$ bsdlabel mirror/gm0e.journal
>> bsdlabel: /dev/mirror/gm0e.journal: no valid label found
>>
>>> And any other .journal you have
>>
>> duncan [202]$ bsdlabel ad8s1a.journal
>> bsdlabel: /dev/ad8s1a.journal: no valid label found
>> duncan [203]$ bsdlabel ad10s1.journal
>> # /dev/ad10s1.journal:
>> 8 partitions:
>> #        size   offset    fstype   [fsize bsize bps/cpg]
>>  a: 1465146065       16    unused        0     0
>>  c: 1465146081        0    unused        0     0         # "raw" part, don't edit
>> partition a: partition extends past end of unit
>> partition c: partition extends past end of unit
>> bsdlabel: partition c doesn't cover the whole unit!
>> bsdlabel: An incorrect partition c may cause problems for standard system utilities
>>
>> Hmm.  That's a bit surprising.  I'm fairly certain that I didn't
>> label ad10s1.journal, but that disk has been used for other
>> things before, and the partition label is the the same (but
>> without the warnings):
>>
>> duncan [204]$ bsdlabel ad10s1
>> # /dev/ad10s1:
>> 8 partitions:
>> #        size   offset    fstype   [fsize bsize bps/cpg]
>>  a: 1465146065       16    unused        0     0
>>  c: 1465146081        0    unused        0     0         # "raw" part, don't edit
>>
>> I guess I should have nuked the starting sectors a bit more
>> thoroughly?
>>
>> That isn't the disk that's giving me panics though.
>>
>>> Just want to double check something.
>>
>> Hope that helps!
>>
>> Cheers,
>>
>> --
>> Andrew
>>
>>
>
> Sorry, i forgot the gm0e and ad8s1a is a label.
> can you show me the bsdlabel for the mirror/gm0 and ad8s1
>
> Also there is a overflow with the ad10s1 and journal
>
> Geom name: gjournal 3015273731
> ID: 3015273731
> Providers:
> 1. Name: ad10s1.journal
>  Mediasize: 749081051136 (698G)
>  Sectorsize: 512
>  Mode: r1w1e1
> Consumers:
> 1. Name: ad10s1
>  Mediasize: 750154793472 (699G)
>  Sectorsize: 512
>  Mode: r1w1e1
>  Jend: 750154792960
>  Jstart: 749081051136
>  Role: Data,Journal
>
> duncan [204]$ bsdlabel ad10s1
> # /dev/ad10s1:
> 8 partitions:
> #        size   offset    fstype   [fsize bsize bps/cpg]
>  a: 1465146065       16    unused        0     0
>  c: 1465146081        0    unused        0     0         # "raw" part,
> don't edit
> partition a: partition extends past end of unit
> partition c: partition extends past end of unit
> bsdlabel: partition c doesn't cover the whole unit!
> bsdlabel: An incorrect partition c may cause problems for standard
> system utilities
>
> When i first started, it kept complaining. For slices, you have the
> ad10, put the s1 slice in, then you gjournal that, bsdlabel the
> .journal. Then change the label to the mediasize of the journal/512.
>
> So in your case your bsdlabel should be 749081051136 / 512 =
> 1463048928 with offset of 0.
> c: 1465146081 - (what is expected)1463048928 = 2097153*512 =
> 1073742336 (which is roughly 1GB, your journal size)
> Your c label is over by about 1GB  which means the fs is writing over
> the journal part sometimes and the journal writing over the data
> sometimes, which would lead to the journal overflow and journal/data
> corruption.
>
> I hope i haven't confused you. I started to gjournal the slices as
> gjournalling the bsdlabels you needed to decrease your bsdlabel by 1
> (thats where the GEOM data was stored) and gave me too many headaches.
>
> If you gjournal the slices, then bsdlabel them, you just change the c
> label to (.journal media size)/512 offset 0, and it all works.
>
> Regards
> David N
>

Could you also post a df (without the -h), if you haven't already
solved the problem.

Cheers
David N


More information about the freebsd-fs mailing list