duplicity-talk
[Top][All Lists]
Advanced

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

Re: [Duplicity-talk] duplicity 0.7 slowness


From: edgar . soldin
Subject: Re: [Duplicity-talk] duplicity 0.7 slowness
Date: Fri, 29 Jul 2016 15:38:22 +0200
User-agent: Mozilla/5.0 (Windows NT 5.1; rv:45.0) Gecko/20100101 Thunderbird/45.2.0

On 29.07.2016 15:36, edgar.soldin--- via Duplicity-talk wrote:
> On 29.07.2016 15:31, address@hidden wrote:
>> edgar.soldin--- via Duplicity-talk <address@hidden> wrote:
>>
>>> On 29.07.2016 00:25, address@hidden wrote:
>>>> edgar.soldin--- via Duplicity-talk <address@hidden> wrote:
>>>>
>>>>> On 25.07.2016 13:02, Philip Jocks via Duplicity-talk wrote:
>>>>>> Hej Aaron,
>>>>>>
>>>>>>> Am 25.07.2016 um 12:46 schrieb Aaron <address@hidden>:
>>>>>>>
>>>>>>> Hello Philip,
>>>>>>>
>>>>>>> On 2016-07-25 10:56, Philip Jocks wrote:
>>>>>>>> Hej Aaron,
>>>>>>>>>> Given your follow-up email, it would be good to rule out SSH as a 
>>>>>>>>>> cause. Can you please backup to a local folder (e.g. 
>>>>>>>>>> file:///tmp/dup_test) with both and see if there is still the time 
>>>>>>>>>> difference?
>>>>>>> [...]
>>>>>>>>> The command I run is
>>>>>>>>> duplicity collection-status --archive-dir '/var/.duply-cache' --name 
>>>>>>>>> duply_zzz_local --encrypt-key XXXXXXXX --encrypt-key YYYYYYYY 
>>>>>>>>> --sign-key XXXXXXXX --verbosity '9' --gpg-options 
>>>>>>>>> '--pinentry-mode=loopback --compress-algo=bzip2 
>>>>>>>>> --bzip2-compress-level=9' --full-if-older-than 1W --volsize 200 
>>>>>>>>> 'file:///path/to/zzz_local'
>>>>>>>>> time’s output:
>>>>>>>>> 0.6: 0.41 real         0.28 user         0.03 sys
>>>>>>>>> 0.7: 71.01 real        37.57 user        23.71 sys
>>>>>>>>> It’s a single chain, full backup has 3 volumes, one incremental with 
>>>>>>>>> 1 volume and one incremental with 317 volumes.
>>>>>>>> have you been able to look into this or maybe even reproduce it?
>>>>>>>
>>>>>>> Apologies, I completely forgot about this -- I'm better with bugs or 
>>>>>>> Launchpad Answers.
>>>>>>>
>>>>>>> Can you please try dropping off all of your additional options and see 
>>>>>>> if the difference persists? I.e., what about:
>>>>>>> duplicity collection-status 'file:///path/to/zzz_local'
>>>>>>> with both versions (or whatever the minimum is that you need to make it 
>>>>>>> run -- I don't use collection-status much)? If that doesn't have the 
>>>>>>> difference, can you please put your options back in one by one to see 
>>>>>>> which looks like it is causing the issue?
>>>>>>
>>>>>> thanks for getting back to me. Haven’t thought about removing options, 
>>>>>> now it’s getting interesting, weirdly. Keeping all options except for 
>>>>>> „—name“ is very fast. Adding „—name duply_zzz_local“ already takes 
>>>>>> several seconds now. So I'd figure, that’s what is at fault?
>>>>>>
>>>>>
>>>>> does it keep being fast after recreating the archive dir?
>>>>>
>>>>> ..ede/duply.net
>>>>>
>>>>> PS: Aaron reworked file selection lately. another user came up w/ a patch 
>>>>> that accelerates duplicity by magnitudes, maybe you want you try it? 
>>>>> https://code.launchpad.net/~mwilck/duplicity/0.7-series/+merge/301332
>>>>>
>>>>
>>>> I tried this patch on my  0.7.07.1 version and the patch failed at the
>>>> first two hunks.  Is that the wrong version to test with?
>>>>
>>>
>>> pretty sure it is against the current bazaar tree. how about editing the 
>>> few lines by hand?
>>
>> Aside from the testing patches which I ommitted, its about 70 lines or
>> so, the first two hunks failed, so its more than just an offset, so I
>> would need some help to make sure I am not doing it incorrectly.
>>
> 
> 
> up to you, maybe you rather wait for the next release then? .. ede/duply.net
> 

btw. you only need to patch globmatch.py and selection.py , the rest is test 
related
  
https://code.launchpad.net/~mwilck/duplicity/0.7-series/+merge/301332/+preview-diff/734481/+files/preview.diff

..ede



reply via email to

[Prev in Thread] Current Thread [Next in Thread]