Re: Add support for fifo files (mkfifo(2)) to syslogd(8)

[christos%zoulas.com@localhost (Christos Zoulas)]

On Nov 8,  6:54pm, anthony.mallet%laas.fr@localhost (Anthony Mallet) wrote:
-- Subject: Re: Add support for fifo files (mkfifo(2)) to syslogd(8)

| Yes, it makes sense to not drop them and queue them instead. Thanks for the
| feedback. I've updated my patch (see below).

Great.

| However, I fought with unrelated bugs while trying to implement the queueing.
| I think they're trivial enough, but I can of course comment on why I think
| those are bugs.
| 
| So two patches:
|  - the unrelated bugs:
|    http://homepages.laas.fr/mallet/syslogd-queues.diff

The first part of the patch looks fine (the numbers are in human format
so strtol() will barf.

The second part of the patch (the NULL check) looks correct to me.

The third part of the patch (the flipped check) also looks correct, because
you want to find a type that will fit the filled entry.

|  - the complete diff for fifo stuff + previous bugs:
|    http://homepages.laas.fr/mallet/syslogd-fifo.diff
| 
| What do you think?

Looks good. My only question is if we should change the FIFO and PIPE writev's
to writev1's so that we have a better chance to put in complete messages in
the slow reader case.

But otherwise it looks fine to me.

christos