discard curl handles found to be "bad"?

[GoogleCodeExporter]

when a curl handle causes s3fs to return -EIO, perhaps should *not*
recycle/returnToPool the curl handle?

I ran into a case where it appeared that I had a "bad" curl handle in the
pool... Dunno how it got bad but then there was no way to get rid of the
bad handle short of umount/mount.

Original issue reported on code.google.com by rri...@gmail.com on 9 Apr 2008 at 3:42

[GoogleCodeExporter]

These files are not big ones. Sizes usually less than 10MB

I was able to get output from unsuccessful attempt to copy a SWF file (size:
2,605,315 bytes) Packets info were caught with the following command:

tcpdump -i eth0 -s 1500 -A -v host s3.amazonaws.com > info.txt

I found the xml message inside info.txt file:

..... HTTP/1.1 400 Bad Request^M
x-amz-request-id: E94B4261A9DD7418^M
x-amz-id-2: [removed_by_zelvlad]
Content-Type: application/xml^M
Transfer-Encoding: chunked^M
Date: Thu, 29 May 2008 19:04:58 GMT^M
Connection: close^M
Server: AmazonS3^M

...

<?xml version="1.0" encoding="UTF-8"?>
<Error><Code>RequestTimeout</Code>
<Message>Your socket connection to the server was not read from or written to 
within
the timeout period. 
Idle connections will be 
closed.</Message><RequestId>E94B4261A9DD7418</RequestId>
<HostId>[removed_by_zelvlad]</HostId></Error>^M

Could it help to resolve the issue. Any suggestions?

Thank you very much for your help -Vlad

Original comment by rri...@gmail.com on 30 May 2008 at 3:18

[GoogleCodeExporter]

Expect: 100-Continue?!?

http://developer.amazonwebservices.com/connect/thread.jspa?threadID=15283&start=
30&tstart=0

Original comment by rri...@gmail.com on 30 May 2008 at 3:33

[GoogleCodeExporter]

http://forum.jungledisk.com/viewtopic.php?t=8203

Original comment by rri...@gmail.com on 30 May 2008 at 7:18

[GoogleCodeExporter]

this issue should be fixed in r177, though would need more feedback/testing 
before
closing

problem in my_curl_progress, should not be differentiating between first time 
and
second and subsequent times

Original comment by rri...@gmail.com on 11 Aug 2008 at 5:20

[GoogleCodeExporter]

Has this issue been resolved? That is, has sufficient testing been done to 
close this one out?

Original comment by dmoore4...@gmail.com on 29 Oct 2010 at 4:12

[GoogleCodeExporter]

Assumed to be fixed.

Original comment by dmoore4...@gmail.com on 19 Dec 2010 at 1:44

• Changed state: Fixed