Re: some procps--n4-4.0.0 fixes

"Dr. Werner Fink" <[email protected]> Fri, 8 Apr 2022 14:49:05 +0200

Newsgroups	gmane.linux.procps.devel
Message-ID	<[email protected]>

On 2022/04/04 01:44:08 -0500, Jim Warner wrote:
> On 4/1/22 1:27 PM, Jim Warner wrote:
> > The addition of elogind to NEWS reminded me to also add the ps stuff.
> > 
> > If you're in the mood, attached is a newlib patch.
> 
> Hi Craig,
> 
> Since the ps AIX behavior was fixed, I couldn't stop fiddling with the
> related ps code.  It's amazing how buggy it still was.
> 
> The attached patches add some much needed robustness.
> 
> In the case of newlib, this one goes on top of the pending NEWS patch.
> 
> Regards,

Hi,

I've a bug report due to new behaviour of sysctl which is a regression

| Hi again,
| after looking a bit more here, I've noticed the issue is caused
| because "sysctl" seems to have changed the output when running:
| 
| Before (and what is expected by Salt would be something like):
| 
| # sysctl -w net.ipv4.ip_forward=1
| net.ipv4.ip_forward = 1
| 
| The actual output we have now (according to Salt minion logs):
| 
| # sysctl -w net.ipv4.ip_forward=1
| /proc/sys/net/ipv4/ip_forward = 1
| 
| This change on the output from "sysctl" is breaking the output
| parsing from Salt, and therefore the state is failing.
| 
| So, this seems to be a valid bug for Salt, unless the change on
| the "sysctl" output was not done on purpose. I have not yet
| clarified when this change was introduced to systemd.

... it is not systemd but procps 4.0.0, nevertheless it is a change
... is this related to

    * sysctl: Check resolved path to be under /proc/sys      issue #179

from NEWS?

Werner

-- 
  "Having a smoking section in a restaurant is like having
          a peeing section in a swimming pool." -- Edward Burr

signature.asc (application/pgp-signature, 894 B)