[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Procmail and mh question



I think it probably is redundant. However, each file name is an integer. 
There does need to be some protection to prevent incoming email from a race 
condition although that is unlikely, but an mh user can easily transfer the 
current message from one mailbox to another. If that occurs at the same 
time as an incoming message to the same mailbox, you might get 2 processes 
storing file 2. But, I am not 100% sure if rcvstore has its own locking 
mechanism.
robert wrote:
>  > From: Jerry Feldman <http://www.blu.org/~gaf>
>  > Date: Tue, 25 Jul 2000 07:26:14 -0400
>  >
>  > In my .procmailrc, for each entry I am locking a lockfile. Since I use MH,
>  
>  > I was wondering if the lock is necessary since rcvstore should do it. Here
>  
>  > is an example of my filter
>  > 	:0 w: BLU/incoming/$LOCKEXT 
>  > 	* ^http://www.blu.org/~TO_discuss
>  > 	| ${RCVSTORE} +BLU/incoming
>  > 
>  > Just curious.
>  > -- 
>  > Jerry Feldman <http://www.blu.org/~gaf>
>  > Boston Linux and Unix user group
>  > http://www.blu.org
> 
> With this recipe, the lock file would be BLU/incoming/.lock, which is not
> terribly useful (i.e., you're locking the entire directory).  I don't use
> mh, but I assume that rcvstore puts the message into a unique file in the
> directory specified.  If so, I would say that the locking was indeed
> redundant.
> 
> -- 
> 







Why do you want this page removed?