I just noticed that the RLimitNPROC is overly agressive when checking the number of processes forked off for a given user. Take the following senario for example: User foo RLimitNPROC 5 5 The user 'foo' has 4 ssh sessions (or 4 of any process) active. If the 'foo' has a cgi that does not fork any additional processes, he can have five concurrent instances of it running as one would expect. However, if the has a cgi that forks even 1 single process, the following error is generated in the logs: test.sh: fork: Resource temporarily unavailable If 'foo' has only 3 ssh sessions, one instance of the script runs fine, but the second will fail when attempting to fork. Based on the documentation, this should not be happening, but it is in the newest 1.3 and 2.0 versions of apache. Has any one else ran across this? --------------------------------------------------------------------- The official User-To-User support forum of the Apache HTTP Server Project. See <URL:http://httpd.apache.org/userslist.html> for more info. To unsubscribe, e-mail: users-unsubscribe@xxxxxxxxxxxxxxxx " from the digest: users-digest-unsubscribe@xxxxxxxxxxxxxxxx For additional commands, e-mail: users-help@xxxxxxxxxxxxxxxx
|