See merge request nexedi/slapos!1402

If there has a stale postmaster.pid, it will be an error if we are
trying to connect the server. In this case, delete it and then start the
service again.

Without this fix, we cannot compile re6st-node because we don't use the
correct version of setuptools.

without the fix, we get this on Debian 12 compilation:

[2023-06-13 16:19:14,522] INFO     Cloning into '/opt/slapgrid/575c4158f610d2eb6886264b60654ce8/parts/shellinabox-git-repository'...
[2023-06-13 16:19:14,526] INFO     /opt/slapgrid/shared/git/e7d6bd941dfb709752c59e8f2a2b9127/libexec/git-core/git-remote-https: error while loading shared libraries: libcurl.so.4: cannot open shared object file: No such file or directory
[2023-06-13 16:19:14,551] INFO     While:
[2023-06-13 16:19:14,551] INFO       Installing shellinabox-git-repository.

See merge request !1401

  Update logs directive to set LogFormat properly

Also switch to basic authentication, this is generally more supported
than digest; keeweb for example only supports basic authentication. It
seems less secure though ( https://github.com/sigoden/dufs/issues/228 )

See merge request nexedi/slapos!1398

Since surykatka 0.8.0 the DB is not updated by default on elapsed time change,
and that shall be user explicit configuration. Thus fix
"software/monitor: Version up surykatka 0.8.0" by adding, required from
edgetest-basic perspective, switch ELAPSED_FAST.

neo logs to sqlite databases, not to "append only" files, so we need extra
care when rotation and cannot use the new "copytruncate" approach.

needs nexedi/erp5!1786

See merge request nexedi/slapos!1395

log files used by neo storage are not text files, they are sqlite
databases, so we can not use copytruncate because it leaves a broken
database. For neo, we have to go back to sending a SIGUSR1 signal to
tell neo client to reopen log files. This depends on neo registering
a signal handler, which is done in [1]. It depends on ZServer, so this
approach will probably have to be adjusted when running on python3
because the current plan is that we don't have ZServer installed on
python3.

This depends on a recent erp5.git where runwsgi understands --pidfile
argument.

1: https://lab.nexedi.com/nexedi/neoppod/blob/fd87e153/neo/client/app.py#L58

In 0409f691 (zope4: Handle log rotation following Zope documentation.,
2022-03-10) we added an option to use copytruncate in logrotate, with
a default value of "false" which is true in jinja2, so it was actually
enabled every where.

This change brings the expected behavior copytruncate option in
logrotate stack, it is not enabled by default, unless explicitly using
copytruncate = true

This can be helpful when comparing measurements or more generally when
trying multiple versions

this is a wrapper independent from where the part got installed

no reason to have an exception here.