Dustin:

Your fix did solve our "hanging" problem - thanks!

As usual, one thing leads to another...

We're now seeing "broken pipe" messages like the following
from the amcheckdump process:

Validating image merlin:/etc datestamp 20100731124607 level 1 part 1/1 on tape 
HBK1_14 file #1
Image was not successfully validated: Error writing data to validation command: 
Error writing fd 9: Broken pipe.
Validating image merlin:/LOCAL datestamp 20100731124607 level 1 part 1/1 on 
tape HBK1_14 file #2
Validating image thor1:/net/ghoss/etc datestamp 20100731124607 level 1 part 1/1 
on tape HBK1_14 file #3
Image was not successfully validated: Error writing data to validation command: 
Error writing fd 9: Broken pipe.

Note that files 1 and 3 fail, but file 2 is okay.

We've written a small script to 'dd if=/dev/rmt/0n bs=32k skip=1 | gtar -tvf -'
through all of the files on the tape and they all seem okay to us.

Since this error is coming from just a few lines after the line
that caused the hanging problem, and since amcheckdump should only
be writing to the /dev/null device that wasn't being opened properly
before, we thought that the problems might be related.

Any ideas would be welcome.

Thanks,



Sean Walmsley / Glenn Cantello

>It's the weirdest thing, but many of Amanda's best bugs (in the sense
>of being the most complex) are discovered independently by 3-4 people
>within a week of one another.
>
>Stephen Gelman, building Amanda on Nextenta, discovered the same bug.
>What's happening is that amfetchdump is using the wrong syntax for
>Perl's open3(), and while it inexplicably works fine on Linux, it does
>not work with the Solaris kernel/libc.
>
>The fix is here:
>  
> http://github.com/zmanda/amanda/commit/bd7108a95c6b21ec5b273c44abc355b179e0795b
>
>you may also be interested in
>  
> http://github.com/zmanda/amanda/commit/438fd709c83e1e7cec91fdd82d5f2d05ec086107
>which fixes a bug in file-descriptor handling in
>Amanda::Xfer::Filter::Process that affects amidxtaped and amfetchdump
>recoveries from compressed or encrypted dumps.
>
>Dustin


=================================================================
Sean Walmsley        sean at fpp . nuclearsafetysolutions dot com
Nuclear Safety Solutions Ltd.  416-592-4608 (V)  416-592-5528 (F)
700 University Ave M/S H04 J19, Toronto, Ontario, M5G 1X6, CANADA

Reply via email to