On Monday, August 1, 2016 at 10:32:55 AM UTC-4, mpl wrote:
>
> On 1 August 2016 at 04:42,  <[email protected] <javascript:>> wrote: 
> > Two quick (I hope) items: 
> > 
> > 1) This is probably config related, but camtool sync --all doesn't do 
> > anything for me... I can do a camtool sync --src localhost --dest server 
> and 
> > it runs fine (or swap src and dest machines)... I thought "all" would 
> just 
> > do both for me??? 
>
> As Eric said, "-all" is about syncing everything that can be synced 
> for a given server (e.g. from the blob storage to the indexer), it's 
> not about syncing from one server to the other. 
>

OK, thanks for the explanation... sorry, I've read the docs and help on 
"all", but I didn't comprehend that it was talking about for a given 
server. I thought it was talking about all servers defined in the 
client-config.json...
 

>
> > 2) When syncing, I get an occasional failure on a blob - 
> > 
> > 2016/07/31 22:37:24 Upload of 
> sha1-dd6bc1cf86d79858e6cc43a037054d99ffdee3e6 
> > to destination blobserver failed: Server didn't receive blob. 
> > 
> > For every one I've ever checked, if I got to the opposite machine and 
> search 
> > for ref:sha1-dd6bc1cf86d79858e6cc43a037054d99ffdee3e6 I find valid data 
> that 
> > appears to match the local machine. 
> > 
> > Is this just a "safe to ignore" item, or...??? 
>
> There are some instances (usually related to the indexer) where things 
> are retried afterwards, which would explain why in the end you don't 
> seem to notice any missing blob. There's also that particular bug: 
> https://github.com/camlistore/camlistore/issues/681 
> But usually, no, it isn't safe to ignore that error, or it should be 
> obvious whether it can be ignored. But we're not quite there yet I 
> think :/ 
> Could you try to check if, when you see such an error message, the 
> same blobRef is mentioned again in the logs afterwards? Maybe that 
> would give us more context to work from. 
>

I had a log from last night that had 15 failures with this error. They all 
appeared only the one time:

Destination needs blob: [sha1-b4408cbc43c47066e30f7a11a9aacd5d6d24d269; 
1562 bytes]
2016/07/31 22:36:06 Upload 
of sha1-b4408cbc43c47066e30f7a11a9aacd5d6d24d269 to destination blobserver 
failed: Server didn't receive blob.

I checked all of them manually on the receiving end (in the GUI doing a 
ref:sha1-b4408cbc43c47066e30f7a11a9aacd5d6d24d269) and the data appeared to 
be there.

On the server side, it stated:

2016/07/31 22:38:23 [SyncHandler /bs/ -> /index/] error copying 
sha1-8cf87cb3e561742d2f9fe330719c641e1509c73d: dest write: unexpected EOF

(yes, that's for a different sha-1, just the error I had handy)

These have been blob-packed stuff (files) I think... so that bug may be the 
cause.

I wonder if I would be advised to rsync between the two stores...


> > Thanks, 
> > Bruce 
> > 
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups 
> > "Camlistore" group. 
> > To unsubscribe from this group and stop receiving emails from it, send 
> an 
> > email to [email protected] <javascript:>. 
> > For more options, visit https://groups.google.com/d/optout. 
>

-- 
You received this message because you are subscribed to the Google Groups 
"Camlistore" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to