Done some work on tonight, mainly some merge requests and bug fixes.

Matching NFS still can cause problems (they changed the way NFS "looks" in the proc filesystem) but I merged in some changes so it hangs less, I hope.

peekd will also work with ARM64 CPUs after someone gave me a patch for that.

I was looking at the net-snmp code trying to work out the differences between the PID file generating code for snmpd and snmptrapd.

πŸ”Ή snmpd uses open() with permissions of 0600
πŸ”Ή snmptrapd uses fopen() with permissions of 0644

Given there on my system there is only one other PID file with 0600 I patched snmpd to use 0644.

If you are running stable on the security update has made it to the repositories. The deb9u5 version is the backport of the 5.0.1 fixes.

That should mean you're ok for most current WordPress security bugs except for CVE-2019-8943 which is a path traversal bug. I think it's fixed for WordPress proper but many modules are still vulnerable.

which finds processes based on your selection criteria and part of now lets you select processes based on state.

Now if you want to go zombie hunting on your server, you can!

I've been triaging a bunch of net-snmp bugs on the bug tracker tonight. There are a lot of old ones for versions of net-snmp long gone.

Strangely there are old bugs that are still there, like why are the permissions for snmp and snmptrapd pid files different?

Clearing out the old ones means I can concentrate on what is left.

I've updated the git repository for -snmp to the latest which is 5.8

The Debian specific patches are now applying cleanly so the next steps are
* to make sure it compiles
* Fix it when it surely won't
* Look at the Debian bugs and fix/close those
* Release!

It's a pretty big codebase that takes forever to compile so it won't be a fast process.

I'm not sure if I read that right but did someone try using lower case signal names in ? And it used to work?

It uses strcmp() which means case sensitive string comparison so how?

Working on , never knew that the various programs use different places to find the length of a command name and also a maximum path.

Hopefully those recent changes have made it more consistent.

I just realised my intro went when my instance crashed all those months ago.

Anyhow, hello (again) I am developer and a programmer including and I write stuff in and

Besides my interests include bike riding, or and reading, mostly There is also a starting interest in

My day job is a senior working on some very large and strange computer networks.

version 3.3.14 is released.

This fixes some pgrep issues where it crashed if it found no matches and you didn't specify a process name.
The pgrep change that matched only on your namespace has also reverted.


Looks like I broke "ps -u" in
On the bright side, I have another test case to add to the testsuite

Procps-ng version 3.3.13 just got released. I have tagged and signed
the last commit and pushed it to gitlab. I will soon start on the
Debian packages and other distributions are free to update as well.

Thankyou to all the contributors!

You can get procps-ng 3.3.13 from the following places:

net-snmp packages have just been uploaded to the ftp-master. They should be available on all mirrors shortly.

This is my first upload of the package after they became orphaned (no maintainer) and have a lot of fixes.

I am considering dropping the bindings. The pysnmp python module is better in almost all ways. There doesn't seem much interest in the net-snmp python bindings anyhow.

vim-command-t package 5.0.2 just uploaded to the FTP master.

This add-on for makes it easy to open files with some clever searching shortcuts.

Well, python-mastodon_1.2.2-1_all.deb built properly now with no lintian warnings.

Still needs some testing but should be ready for uploading soon.

Started working on packaging @halcy module for

Looks like I'll need to package pytest-vcr first or skip the tests.

has now gone to v3.3.13rc1

This will be the version used by the translators to update the language files. No more code updates are expected unless there is a problem found.

Been spending some time on the net-snmp bugs in the bug tracker. There are some very very old (10 years) bugs there that I have closed as they appear to be upstream ones and there have been many releases since.

That means I can concentrate on the remainder much easier.

Show more
Mastodon on Dropbear

Welcome to my small corner of the Mastodon fediverse.