Lars Marowsky-Bree wrote:
On 2006-01-26T07:36:17, Alan Robertson <[EMAIL PROTECTED]> wrote:
I decided fixing it would probably let me sleep better.
You can retry with the latest CVS HEAD, I think it should work better
now, I think it should solve this issue for you. Actually you only need
to replace hto-mapfuncs with one generated from configure from that
build.
Actually, the _right_ fix is to change the LRM to redirect stderr to
stdout and then get rid of some code in the LRM.
However, ResourceManager.in also expects the status to come out on
stdout and ignores stderr, so to make the scripts work in hb1 mode too,
I choose this path.
Besides, I didn't want to touch the LRM and was edgy because this bug
essentially broke hb1-resource scripts _completely_ and thus can't ever
have been tested.
But yes, I agree in principle. Turns out the LRM actually already
_reads_ stderr, it just doesn't look at it for such patterns. It also
only expects this in the first line, while the old script would have
basically concatenated all lines. It's not very tolerant of inputs.
In practice, if it actually paid attention to line boundaries, then the
first line would be enough in most or all cases.
I actually meant in a sort of 2>&1 way - so that we could get rid of all
the code that deals with stderr at all...
--
Alan Robertson <[EMAIL PROTECTED]>
"Openness is the foundation and preservative of friendship... Let me
claim from you at all times your undisguised opinions." - William
Wilberforce
_______________________________________________________
Linux-HA-Dev: [email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/
