svn commit: r316938 - head/sbin/savecore

Ngie Cooper (yaneurabeya) yaneurabeya at gmail.com
Fri Apr 14 20:46:52 UTC 2017


> On Apr 14, 2017, at 13:43, Larry Rosenman <ler at lerctr.org> wrote:
> 
> On 4/14/17, 3:39 PM, "Ngie Cooper (yaneurabeya)" <owner-svn-src-all at freebsd.org on behalf of yaneurabeya at gmail.com> wrote:
> 
>> On Apr 14, 2017, at 13:37, Larry Rosenman <ler at lerctr.org> wrote:
>> 
>> On 4/14/17, 3:33 PM, "Ngie Cooper (yaneurabeya)" <yaneurabeya at gmail.com> wrote:
>> 
>>> On Apr 14, 2017, at 13:26, Larry Rosenman <ler at lerctr.org> wrote:
>>> 
>>> On 4/14/17, 3:19 PM, "Ngie Cooper (yaneurabeya)" <owner-svn-src-all at freebsd.org on behalf of yaneurabeya at gmail.com> wrote:
>>> 
>>>> On Apr 14, 2017, at 13:14, Slawa Olhovchenkov <slw at zxy.spb.ru> wrote:
>>>> 
>>>> On Fri, Apr 14, 2017 at 01:49:51PM -0600, Alan Somers wrote:
>>>> 
>>>>> On Fri, Apr 14, 2017 at 1:41 PM, Ngie Cooper <ngie at freebsd.org> wrote:
>>>>>> Author: ngie
>>>>>> Date: Fri Apr 14 19:41:48 2017
>>>>>> New Revision: 316938
>>>>>> URL: https://svnweb.freebsd.org/changeset/base/316938
>>>>>> 
>>>>>> Log:
>>>>>> savecore: fix space calculation with respect to `minfree` in check_space(..)
>>>>>> 
>>>>>> - Use strtoll(3) instead of atoi(3), because atoi(3) limits the
>>>>>> representable data to INT_MAX. Check the values received from
>>>>>> strtoll(3), trimming trailing whitespace off the end to maintain
>>>>>> POLA.
>>>>>> - Use `KiB` instead of `kB` when describing free space, total space,
>>>>>> etc. I am now fully aware of `KiB` being the IEC standard for 1024
>>>>>> bytes and `kB` being the IEC standard for 1000 bytes.
>>>>>> - Store available number of KiB in `available` so it can be more
>>>>>> easily queried and compared to ensure that there are enough KiB to
>>>>>> store the dump image on disk.
>>>>>> - Print out the reserved space on disk, per `minfree`, so end-users
>>>>>> can troubleshoot why check_space(..) is reporting that there isn't
>>>>>> enough free space.
>>>>>> 
>>>>>> MFC after:    7 weeks
>>>>>> Reviewed by:  Anton Rang <rang at acm.com> (earlier diff), cem (earlier diff)
>>>>>> Tested with:  positive/negative cases (see review); make tinderbox
>>>>>> Sponsored by: Dell EMC Isilon
>>>>>> Differential Revision:        D10379
>>>>> 
>>>>> The free space calculation is still uselessly conservative, because it
>>>>> doesn't account for the fact that core dumps will always be either
>>>>> spare or compressed.  The result is that savecore will frequently
>>>>> refuse to save corefiles even when there's plenty of space.  I
>>>>> proposed removing the space check altogether in
>>>>> https://reviews.freebsd.org/D2587.  However, I agreed to wait until
>>>>> after the compressed core dump feature was merged, because then mostly
>>>>> accurate space checks will be possible.  AFAIK the compressed core
>>>>> dump feature still hasn't been finished.
>>>> 
>>>> Is posible (in the future) to use multiple swaps (on multiple disks)
>>>> for save core dumps?
>>> 
>>>  Multiple swap devices is already handled by savecore(8), if one uses fstab(5) or dumpon(8). Otherwise, you must invoke savecore(8) on individual devices.
>>> 
>>>  As far as saving to multiple disks is concerned, I would hope that one is using a redundancy capable filesystem (zfs) or RAID-like technology (gmirror, graid, LSI Fusion’s RAID product line) to stripe and/or mirror the data across multiple disks.
>> 
>>>> 
>>> How do I use multiple devices to have the system dump on all of my swap?  I got a message about not enough space, but there (I think) was enough between multiple drives….
>> 
>>   Something like:
>> 
>>   - Create a zpool
>>   - Mount zpool to /crashdumps
>>   - Change dumpdir in /etc/rc.conf to be /crashdumps, e.g., echo ‘dumpdir=/crashdumps’
>> 
>>   ?
>>   HTH,
>>   -Ngie
>> 
>>   PS The issue with lack of space might be the issue that Alan brought up earlier with compressed dumps and overly conservative free space checks, or it might be the fact that dumpdir (default: /var/crash) is full.
>> 
>> 
>> I was talking about the actual crashdump to swap by the system.  /var/crash has 10T of space (my root pool).
> 
>    If your memory is bigger than your swap, you’re unfortunately not able to save the mini dump if the size of the saved pages exceed the space on swap. I think this is where markj’s compressed dumps feature will come in handy.
>    Thanks!
>    -Ngie
> 
> 
> Yeah, I have the following:
> borg.lerctr.org /home/ler $ swapctl -l
> Device:       1024-blocks     Used:
> /dev/mfid0p3    8388608         0
> /dev/mfid1p3    8388608         0
> /dev/mfid2p3    8388608         0
> /dev/mfid3p3    8388608         0
> /dev/mfid4p3    8388608         0
> /dev/mfid5p3    8388608         0
> borg.lerctr.org /home/ler $ sysctl hw.physmem
> hw.physmem: 137368682496
> borg.lerctr.org /home/ler $
> 
> SO 6 8G partitions (48G), but the dump is larger than 8G.

Yup, that’s not going to work. Since you’re using mfi(4), I’d try creating another 48GB large RAID-0 volume to help deal with that issue.
Cheers,
-Ngie
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 842 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://lists.freebsd.org/pipermail/svn-src-head/attachments/20170414/2a42b877/attachment.sig>


More information about the svn-src-head mailing list