|
Hello,
I have one more doubt about sieve scripts. I thought that sieve scripts would be sync on replica at the same time that mailbox are, but I haven't check it yet. Now I see that sieve is not sync, and I'm not sure if there is a automatic way to do it. How can i sync sieve at the same time that mailbox? Or I should to do it manually? Regards and thaks for all your support! Manel Gimeno Zaragoza magiza83@xxxxxxxxxxx > From: brong@xxxxxxxxxxx > To: magiza83@xxxxxxxxxxx > CC: info-cyrus@xxxxxxxxxxxxxxxxxxxx; eric.luyten@xxxxxxxxx > Subject: RE: Problem recover replica > Date: Thu, 23 Feb 2012 19:50:03 +0100 > > Well found :) The rfc822. stuff is calculated on the fly, for testing spool against index file. > > On Thu, Feb 23, 2012, at 03:45 PM, Manel Gimeno Zaragozá wrote: > > > > Hello, > > > > I've found the problem by my self, I was asking the wrong attribute. The correct one is "digest.sha1": > > > > my $s1Size = $IMAPs1->fetch('1:*', 'digest.sha1'); > > > > Now its almost instantaneous: > > > > real 0m0.395s > > user 0m0.368s > > sys 0m0.016s > > > > and its perfect to compare the integrity of the replica! > > > > Thanks & Regards! > > > > Manel Gimeno Zaragoza > > magiza83@xxxxxxxxxxx > > > > > > From: magiza83@xxxxxxxxxxx > > To: brong@xxxxxxxxxxx > > Subject: RE: Problem recover replica > > Date: Thu, 23 Feb 2012 13:44:32 +0100 > > CC: info-cyrus@xxxxxxxxxxxxxxxxxxxx; eric.luyten@xxxxxxxxx > > > > > > > > > > > > > > > > Hello Bron, > > > > I've apply the patch on cyrus 2.4.13 succesfully and now I can fetch "rfc822.sha1" and "rfc822.filesize": > > > > . fetch 1:* rfc822.sha1 > > * 1 FETCH (RFC822.SHA1 284c8e2602d2094609aefb31c64622dbcc8a59cb) > > * 2 FETCH (RFC822.SHA1 40d4f8a1a1c214385d875181c7e950ea04571855) > > * 3 FETCH (RFC822.SHA1 756b897109ba58dc78c451ffa1ba6837ec17a391) > > > > . fetch 1:* rfc822.filesize > > * 1 FETCH (RFC822.FILESIZE 1402) > > * 2 FETCH (RFC822.FILESIZE 1399) > > * 3 FETCH (RFC822.FILESIZE 2097759) > > > > But I have some doubts about it. Fetch sha1 takes too much, I have 1500 mail (around 3GB of data) in this account and it take over 20 seconds to do it: > > real 0m20.288s > > user 0m0.380s > > sys 0m0.014s > > > > On the other hand fetch filesize takes 0,4 secs: > > real 0m0.401s > > user 0m0.369s > > sys 0m0.022s > > > > I thought that sha1 was already the GUID field and fetch them would be really fast, but it isn't. have I made something wrong? My perl code is the same in both "fetch": > > > > my $s1Size = $IMAPs1->fetch('1:*', 'rfc822.filesize'); > > > > Finally, I attach the patch code for 2.4.13. As you said I had to made some "minor" changes (some compile error due to some missing object) and find the correct lines to apply it. > > > > Thanks & Regards. > > > > Manel Gimeno Zaragoza > > magiza83@xxxxxxxxxxx > > > > > > > Date: Wed, 22 Feb 2012 17:40:48 +0100 > > > From: brong@xxxxxxxxxxx > > > To: brong@xxxxxxxxxxx > > > CC: magiza83@xxxxxxxxxxx; eric.luyten@xxxxxxxxx; info-cyrus@xxxxxxxxxxxxxxxxxxxx > > > Subject: Re: Problem recover replica > > > > > > On Wed, Feb 22, 2012 at 03:49:44PM +0100, Bron Gondwana wrote: > > > > Ahh, you need our guid commands patch! > > > > > > > > It's on the fastmail branch on github. Can email a direct link when I'm not on my phone... > > > > > > https://github.com/brong/cyrus-imapd/commit/e0c404436c69bf9535c7f86e7e739c16c6064399 > > > > > > That's one that applies directly on top of master... you > > > may need to fiddle with some fuzz to get it to apply to > > > 2.4. There's lots of extra annotation stuff on master > > > now. > > > > > > Bron. > > > > > > ---- > > Cyrus Home Page: http://www.cyrusimap.org/ > > List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/ > -- > Bron Gondwana > brong@xxxxxxxxxxx > |
---- Cyrus Home Page: http://www.cyrusimap.org/ List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
