Re: [Duplicity-talk] Fwd: Re: Duplicity process killed

[covici]

I am not seeing this particular bug here, backing up to Azure and my
backup directory is now 11g!  My only problem is that duplicity is very
slow, takes about 4 hours to walk the tree even if very little to back
up.  I have not  tried to do a verify, because it took 55 hours to do my
full backup.

Chris Vanden Berghe via Duplicity-talk <address@hidden> wrote:

> Hi Kenneth,
> 
> The data on the server does not change very often. I do a full backup
> once per month and incremental in between. I keep the backup chains for
> 2 months.
> 
> # du -h duplicity-full-signature*.gpg
> 1018M   duplicity-full-signatures.20160715T164934Z.sigtar.gpg
> 
> Regards,
> Chris.
> 
> ---
> On 15/07/2016 19:55, Kenneth Loafman via Duplicity-talk wrote:
> > Jumping in here... How often do you do FULL backups?  Are you building
> > a long chain of incrementals?  What was the size of the last
> > duplicity-full-signature*.gpg file?  
> >
> > On Fri, Jul 15, 2016 at 5:59 AM, <address@hidden
> > <mailto:address@hidden>> wrote:
> >
> >     ok
> >      http://docs.pythonboto.org/en/latest/releasenotes/v2.41.0.html
> >     does not list any memory specific fix.
> >     remember now you already use duplicity 0.7.08 .. so no updates
> >     necessary.
> >
> >     next would probably be patching duplicity to output memory usage
> >     when in max verbosity. don't see anybody doing that unfortunately,
> >     do you offer :) ?  i'll cc Ken the maintainer, maybe he has
> >     another idea?
> >
> >     to make sure it is duplicity and not the backend code, you could
> >     do a backup w/ identical settings to a local file:// target and
> >     see if the second run sports the same symptoms you experience now.
> >
> >     additionally can you send me a _complete_ terminal output of the
> >     killed duplicity run. you may want obfuscate strings in it beforehand.
> >
> >     ..ede/duply.net <http://duply.net>
> >
> >     On 15.07.2016 12:37, edgar.soldin--- via Duplicity-talk wrote:
> >     > ok. before we go further - please make sure you have the latest
> >     duplicity release and latest boto, if not install them locally
> >     (sent you the mini howto earlier).
> >     >
> >     > can you run a verify or does that also die assuming you run a
> >     backup command?
> >     >
> >     > ..ede
> >     >
> >     > On 15.07.2016 11:45, Chris Vanden Berghe via Duplicity-talk wrote:
> >     >> Scrolling down was too obvious, I guess ;-) I have attached a
> >     screenshot
> >     >> of what I see a second before Duplicity is killed.
> >     >>
> >     >> The source files are ~82 GB. The last successful backup:
> >     >>
> >     >> # /usr/bin/duplicity collection-status
> >     --archive-dir=/root/Backup/cache/
> >     >> gs://bobo-backup/
> >     >> Local and Remote metadata are synchronized, no sync needed.
> >     >> Last full backup date: Mon Jul  4 17:11:39 2016
> >     >> Collection Status
> >     >> -----------------
> >     >> Connecting with backend: BackendWrapper
> >     >> Archive dir: /root/Backup/cache/01f81df3eb37282f76c93fb89a3708f5
> >     >>
> >     >> Found 0 secondary backup chains.
> >     >>
> >     >> Found primary backup chain with matching signature chain:
> >     >> -------------------------
> >     >> Chain start time: Mon Jul  4 17:11:39 2016
> >     >> Chain end time: Mon Jul  4 17:11:39 2016
> >     >> Number of contained backup sets: 1
> >     >> Total number of contained volumes: 254
> >     >>  Type of backup set:                            Time:      Num
> >     volumes:
> >     >>                 Full         Mon Jul  4 17:11:39 2016         
> >          254
> >     >> -------------------------
> >     >> No orphaned or incomplete backup sets found.
> >     >>
> >     >> Regards,
> >     >> Chris.
> >     >>
> >     >> ---
> >     >> On 15/07/2016 15:36, edgar.soldin--- via Duplicity-talk wrote:
> >     >>>> On 15.07.2016 05:39, Chris Vanden Berghe via Duplicity-talk
> >     wrote:> Hi Edgar,
> >     >>>>> I was unsuccessful in getting useful data from htop. It
> >     seems like tree
> >     >>>>> view and sorting on memory usage features are mutually
> >     exclusive in
> >     >>>>> htop... so when I enable tree view I no longer see the
> >     duplicity process
> >     >>>>> (off screen) and when I sort on memory I do not see
> >     Duplicity's children
> >     >>>>> (if any).
> >     >>>> hehe, read into how htop works. you can scroll the list w/
> >     the cursor keys.
> >     >>>>
> >     >>> sort by tree _and_ memory are mutually exclusive. so you have
> >     to sort by tree and scroll to duplicity, make the terminal big
> >     enough to see all it's sub processes and look out for the culprit
> >     yourself.
> >     >>>
> >     >>> btw. how big is your backup. size of data, number/size of
> >     volumes per full/chain.
> >     >>>
> >     >>> ..ede
> >     >>>
> >     >>>
> >     >>> _______________________________________________
> >     >>> Duplicity-talk mailing list
> >     >>> address@hidden <mailto:address@hidden>
> >     >>> https://lists.nongnu.org/mailman/listinfo/duplicity-talk
> >     >>
> >     >>
> >     >>
> >     >> _______________________________________________
> >     >> Duplicity-talk mailing list
> >     >> address@hidden <mailto:address@hidden>
> >     >> https://lists.nongnu.org/mailman/listinfo/duplicity-talk
> >     >>
> >     >
> >     > _______________________________________________
> >     > Duplicity-talk mailing list
> >     > address@hidden <mailto:address@hidden>
> >     > https://lists.nongnu.org/mailman/listinfo/duplicity-talk
> >     >
> >
> >
> >
> >
> > _______________________________________________
> > Duplicity-talk mailing list
> > address@hidden
> > https://lists.nongnu.org/mailman/listinfo/duplicity-talk
> 
> 
> _______________________________________________
> Duplicity-talk mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/duplicity-talk

-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

         John Covici
         address@hidden