Re: Clock drift and other open issues: Collecting information
Johnny Billquist <[email protected]>
| Newsgroups | gmane.os.netbsd.ports.vax |
|---|---|
| Message-ID | <[email protected]> |
On 2023-12-30 21:14, Mouse wrote:
>>> That's one reason I tend to use sysctl for querying, and sysctl -w
>>> for changing, resource limits. (The other is that the syntax is
>>> significantly less shell-dependent.)
>> The problem with that is that you can deal with the hard limits, but
>> tricky with process limits. The classic "what commands must be
>> built-in in a shell".
>
> proc.$$.rlimit (rather than proc.curproc.rlimit) is your friend in that
> regard.
Cool. I didn't know you had access to random process rlimits that way. Nice.
> The "what commands must be built-in" applies only when there's no
> interface that lets a separate process affect the settings for the
> shell. When the interface is setrlimit(), that's so, but with sysctl a
> child process _can_ affect the shell's resource limits.
I had missed some details on that one. Thanks.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: [email protected] || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol