Current-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: lvm+DIOCCACHESYNC woes



On 06/06/11 13:59, Adam Hamsik wrote:
> Right now LVM just sends DIOCCACHESYNC to physical device behind
> 
> dm_target_linear.c::dm_target_linear_sync
> 
> return VOP_IOCTL(tlc->pdev->pdev_vnode,  DIOCCACHESYNC, &cmd,
>             FREAD|FWRITE, kauth_cred_get());
> 
> So to me it looks like this error comes from disk device itself.
> 
>> Since that post I've been running various NetBSD domUs in various
>> versions of -current and this issue has not manifested, so I assume a
>> recent change caused this?
>>
>> Or maybe I fscked up somewhere. Feedback welcome :)
>>
>> Hugo
> 
> Regards
> 
> Adam.
> 


Hello Adam,

What is the nature of this error, should I be concerned?

And more importantly, how to make it stop?:D

syslogd[173]: last message repeated 192 times
syslogd[173]: last message repeated 210 times
syslogd[173]: last message repeated 156 times
syslogd[173]: last message repeated 166 times
syslogd[173]: last message repeated 172 times
syslogd[173]: last message repeated 146 times
syslogd[173]: last message repeated 148 times
syslogd[173]: last message repeated 206 times
syslogd[173]: last message repeated 198 times
syslogd[173]: last message repeated 172 times

I realize of course that commenting it out from kernel source and/or
stopping syslogd would do it, but in the interest of other users running
into this later.. wanna dig to the bottom of it?

The disk is:
sd0 at scsibus0 target 0 lun 0: <ATA, WDC WD5003ABYX-0, 1S01> disk fixed
sd0: 465 GB, 476941 cyl, 16 head, 127 sec, 512 bytes/sect x 976773168
sectors

Now that I think about it, all the other dom0s ran on SATA/IDE disks.
Other than differing -current versions (altough I did install another
5.99.52 quite recently, and didn't see this), this is the only other "in
context" difference that comes to mind.


Home | Main Index | Thread Index | Old Index