Hi,

Been trying to find the source of an error for a long while now when 
uploading files using pk-put to a server. Though I'd ask to see if anyone 
had seen somthing similar and had some pointers on where I should look?

Thanks for any pointers and please find description below.

/Viktor


I am trying to bcp an external disk, using:

*pk-put --verbose file --permanode --title="Backup of External 1 
2018-11-17" --tag=backup,bcp ./disk*

This worked fine for another disk earlier today (bcp visible as permanode 
in the web interface and no errors), but now I keep getting the following 
error (retried many times by now..):

*2018/11/17 20:48:07 upload http error: Post 
...bs-and-maybe-also-index/camli/upload: http2: Transport: cannot retry err 
[http2: Transport received Server's graceful shutdown GOAWAY] after 
Request.Body was written; define Request.GetBody to avoid this error*
*2018/11/17 20:48:07 Error uploading map for disk/bcp/20130323/Photo 
Library/2003/Nytt dass till Lyr (directory, 
sha224-19b0cb6c67223a0a8baaf4b9b3afbbd5464f010d57a803617e318b7b): upload 
http error: Post .../bs-and-maybe-also-index/camli/upload: http2: 
Transport: cannot retry err [http2: Transport received Server's graceful 
shutdown GOAWAY] after Request.Body was written; define Request.GetBody to 
avoid this error*
*2018/11/17 20:48:07 Error uploading disk/bcp/20130323/Photo 
Library/2003/Nytt dass till Lyr: upload http error: Post 
.../bs-and-maybe-also-index/camli/upload: http2: Transport: cannot retry 
err [http2: Transport received Server's graceful shutdown GOAWAY] after 
Request.Body was written; define Request.GetBody to avoid this error*

(900 files uploaded successfully before this error occurred, so it does not 
happen straight away)

The perkeepd server that I am uploading to is behind a nginx reverse-proxy, 
however, looking at the nginx error log I see nothing whereas the nginx 
access log shows:

*IPHERE - viktor [17/Nov/2018:20:48:07 +0100] "POST 
/bs-and-maybe-also-index/camli/upload HTTP/2.0" 499 0 "-" 
"Go-http-client/2.0"*
*IPHERE - viktor [17/Nov/2018:20:48:07 +0100] "POST 
/bs-and-maybe-also-index/camli/upload HTTP/2.0" 499 0 "-" 
"Go-http-client/2.0"*

I.e. nginx has logged this as having responded 499 (which represents a 
client error if the http client closes the connection while nginx is 
processing it).
I see no errors in the perkeepd log (see below). I am thus struggling to 
understand if the error is server side or client side, they seem to be 
blaming each other :-)

*Nov 17 20:48:07 box perkeepd[868]: blobpacked: 2018/11/17 20:48:07 Packing 
file sha224-92f719fbf35e0b796272b31db5d10ddbfcee635195b8d9c9575c6774 ...*
*Nov 17 20:48:07 box perkeepd[868]: blobpacked: 2018/11/17 20:48:07 Error 
packing file 
sha224-92f719fbf35e0b796272b31db5d10ddbfcee635195b8d9c9575c6774: already 
have wholeref*
*Nov 17 20:48:07 box perkeepd[868]: blobpacked: 2018/11/17 20:48:07 Packing 
file sha224-b6f705aba652fc73c68bed9c154ce7079ea080b86c50326f0e67992f ...*
*Nov 17 20:48:07 box perkeepd[868]: blobpacked: 2018/11/17 20:48:07 Error 
packing file 
sha224-b6f705aba652fc73c68bed9c154ce7079ea080b86c50326f0e67992f: already 
have wholeref*
*Nov 17 20:48:07 box perkeepd[868]: 2018/11/17 20:48:07 Received blob 
[sha224-92f719fbf35e0b796272b31db5d10ddbfcee635195b8d9c9575c6774; 888 
bytes]*
*Nov 17 20:48:07 box perkeepd[868]: 2018/11/17 20:48:07 Received blob 
[sha224-b6f705aba652fc73c68bed9c154ce7079ea080b86c50326f0e67992f; 1237 
bytes]*

-- 
You received this message because you are subscribed to the Google Groups 
"Perkeep" 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