duplicity-talk
[Top][All Lists]
Advanced

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

Re: [Duplicity-talk] Problems reading/writing to Rackspace Cloud Files (


From: edgar . soldin
Subject: Re: [Duplicity-talk] Problems reading/writing to Rackspace Cloud Files (UK) backend
Date: Fri, 19 Oct 2012 10:32:46 +0200
User-agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:16.0) Gecko/20121010 Thunderbird/16.0.1

On 18.10.2012 20:05, Lou wrote:
> On Thu, 18 Oct 2012 15:34:18 +1100
> Miguel Jacq <address@hidden> wrote:
>  
>> Hi,
>>
>> I've recently (last two weeks or so) been experiencing SSL read/write
>> timeout errors when using Rackspace Cloud Files (the London DC) as the
>> backend, via the python-cloudfiles module [1].
>>
>> See pastie as an example: http://pastie.org/private/gud3nlj67ymf5d1rgkv8da
>>
>>
>> This has happened previously in the past but 'resolved itself'
>> (presumably Rackspace fixed the underlying issue). It's totally random
>> - - sometimes it works, sometimes it doesn't; no rhyme or reason.
>>
>>
>> This time it has been dragging on, so I contacted Rackspace for help,
>> and they've asked if I can provide 'transaction IDs' of my read/write
>> attempts.
>>
>>
>> I upped the verbosity in Duplicity to -v9, but I can't see any
>> transaction IDs.
>>
>> I realise this might not be specific to Duplicity, as it'd be the API
>> that would be able to display the transaction IDs - but has anyone
>> experienced this issue with Duplicity of late, and perhaps knows how
>> to expose this debug information - perhaps in the python-cloudfiles
>> module directly?
>>
>>
>> Thanks for any assistance!
>>
>> Mig
> 
> I also occasionally encounter this (or a similar) issue here in the US.
> Last time was Oct 7, but I only backup once a week. I've never resolved
> it due to the sporadic timing.
> 
> The next backup is on Sunday. I'll increase the verbosity to see if
> there is anything of interest. I'm using python-cloudfiles 1.7.11 and
> Duplicity 0.6.19 on a Debian Squeeze server.
> 
> Unfortunately I'm not sure how to expose these transaction IDs, but
> you're not the only one experiencing this.
> 
> Lou
> 

ok, had a look at the provided pastebin error stack. it pretty clearly states 
it's an http timeout, not a cloudfiles error per se.

i suggest you to raise '--num-retries' (default 5) and keep a close eye on your 
backups output. you'll probably see lines that inform you of timeouts but'll 
retry after 30 seconds.

what's weird is in your paste it states it died after the first try... is that 
correct?


..ede/duply.net



reply via email to

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