Re: [PATCH] pygrub: add syslog support to pygrub

[Zhigang Wang <zhigang.x.wang@oracle.com>]

On 07/19/2012 03:07 PM, Ian Campbell wrote:
> On Thu, 2012-07-19 at 19:19 +0100, Zhigang Wang wrote:
>> # HG changeset patch
>> # User Zhigang Wang <zhigang.x.wang@oracle.com>
>> # Date 1342720736 14400
>> # Node ID ec9655b30a5fa5b5abb3e05505f681f9be559613
>> # Parent  43e21ce7f22151524b800a6cf0ac4ba1233b34a7
>> pygrub: add syslog support to pygrub
>>
>> Currently, when pygrub failed, we don't know the reason because xend/xl will
>> not capture the stderr message.
> Actual xl in unstable will log to /var/log/xen/bootloader.<domid>.log
> (or something like that) and print a message on failure directing the
> user to it.
I see. But it seems the design is wrong: you want all logs in stderr.
>
>> This patch will log the error message to syslog.
> I suppose there is no harm in also (optionally?) logging to syslog as
> well, but I think we want to keep the messages to stdout also.
I think we should: logs to syslog, errors to stderr.
>
>> Also in this patch:
>>
>> 1). Fix indentation for some lines.
>> 2). Removed some trailing spaces.
> These two make up the vast bulk of the changes and really obscure the
> actual functional changes.
>
> Please can you do this cleanup in a separate patch.
>
>> 3). Mark 'isconfig' a duplicate option of 'debug' and remove the currently broken code:
>>
>>       if isconfig:
>>           chosencfg = run_grub(file, entry, fs, incfg["args"])
>>
>>     'fs' is not defined yet here, so it will raise an exception.
> This should also be a separate patch.
>
> I think all of this is 4.3 material and will therefore have to wait
> until 4.2 branches. Please can you resend at that point.
>
> Ian.
I can do it on 4.3.

Thanks,

Zhigang