lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20250402054051.GA22424@linuxonhyperv3.guj3yctzbm1etfxqx2vob5hsef.xx.internal.cloudapp.net>
Date: Tue, 1 Apr 2025 22:40:51 -0700
From: Shradha Gupta <shradhagupta@...ux.microsoft.com>
To: Dexuan Cui <decui@...rosoft.com>
Cc: KY Srinivasan <kys@...rosoft.com>,
	Haiyang Zhang <haiyangz@...rosoft.com>,
	Wei Liu <wei.liu@...nel.org>,
	"linux-hyperv@...r.kernel.org" <linux-hyperv@...r.kernel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Shradha Gupta <shradhagupta@...rosoft.com>
Subject: Re: [PATCH v2] hv/hv_kvp_daemon: Enable debug logs for hv_kvp_daemon

On Tue, Apr 01, 2025 at 07:18:04PM +0000, Dexuan Cui wrote:
> > From: Shradha Gupta <shradhagupta@...ux.microsoft.com>
> > Sent: Monday, March 31, 2025 9:06 PM
> > > > +static void convert_tm_to_string(char *tm_str, size_t tm_str_size)
> > > > +{
> > > > +	struct tm tm;
> > > > +	time_t t;
> > > > +
> > > > +	time(&t);
> > > > +	gmtime_r(&t, &tm);
> > > > +	strftime(tm_str, tm_str_size, "%Y-%m-%dT%H:%M:%S", &tm);
> > > > +}
> > >
> > > Now the function is unnecessary since v2 uses syslog(), which already
> > > prefixes every message with a timestamp.
> > 
> > Hi Dexuan,
> > I have deliberately kept this timestamp in the raw message so that
> > if/whenever they are redirected to other file, irrespective of the
> > configuration of the syslog we have valid timestamp for debugging
> 
> A message produced by syslog() is always prefixed with a timestamp,
> and IMO can't be redirected.  By "redirected to other file",  I guess
> you mean systemd's options StandardOutput= and StandardError=
> for a service, but those are stdout/err, not syslog().

rsyslog can be configured to forward syslog logs from a service to a file.
If the timestamp template in rsyslog.conf is not configured, the timestamps
would be missed in forwarded file.
But I think that would be an issue to be handled by the script/service
forwarding these logs. It can be easily fixed by adding the timestamp
template.

I will remove the timestamp from the raw log message then. Thanks

> 
> > > > +static void kvp_dump_initial_pools(int pool)
> > > > + [...]
> > > > +	for (i = 0; i < kvp_file_info[pool].num_records; i++)
> > > > +		syslog(LOG_DEBUG, "[%s]: pool: %d, %d/%d key=%s
> > > > val=%s\n",
> > > > +		       tm_str, pool, i, kvp_file_info[pool].num_records,
> > >
> > > Can you change the 'i' to 'i+1'? This makes the messages a little more
> > > natural to users who are not programmers :-)
> > sure, but I am just worried that might cause confusion when someone
> > tried to co-relate it with the actual kv_pool_{i} contents that start
> > with 0.
> IMO these messages are mostly for admins, who would feel more
> natural when seeing N/N as the last element, compared  with N-1/N.
Got it, will modify this too
> 
> Thanks,
> Dexuan

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ