(
Jump to the original submission)
Sun 13 Dec 2015 12:56:01 AM UTC, comment #6:
Fixed in [09dcafc].
|
Sat 12 Dec 2015 10:53:52 PM UTC, comment #5:
Short example:
seq 1 30003| ppar --retries 2 --block 100k -S a.a,: --pipe 'wc;hostname'
|
Sun 25 Oct 2015 08:19:05 AM UTC, comment #4:
Can we reuse $job->{'stdin_buffer'}
|
Sun 25 Oct 2015 06:43:56 AM UTC, comment #3:
One way:
if --retry:
# store a copy of the block before fork.
$job->set_block(substr($$record_ref,$endpos,length $$record_ref));
|
Sun 25 Oct 2015 06:32:33 AM UTC, comment #2:
Data is given to write_record_to_pipe, which forks and writes to pipe and exits.
When returning from write_record_to_pipe (but before data is written) the buffer is shortened.
Some redesign is needed before this bug can be fixed.
|
Sun 07 Jun 2015 10:39:07 PM UTC, comment #1:
Better example:
seq 1 300030| parallel --retries 2 -S no.server,: --pipe 'wc;hostname'
It should run everything on : when retrying.
|
Tue 05 May 2015 02:58:29 PM UTC, original submission:
seq 1 300030| parallel --retries 2 -S redhat9.t,: --pipe 'wc;hostname'
Re-schedules on redhat9.t and not on :.
|
(Note: upload size limit is set to 16384 kB, after insertion of the required
escape characters.)
Attach File(s):
Comment:
No files currently attached
Depends on the following items: None found
Items that depend on this one: None found
Follow 2 latest changes.