AF:
NF:0
PS:10
SRH:1
SFN:
DSR:
MID:<20070924162427.70fe3488@ripper.onstor.net>
CFG:
PT:0
S:andy.sharp@onstor.com
RQ:
SSV:onstor-exch02.onstor.net
NSV:
SSH:
R:<larry.scheer@onstor.com>
MAID:1
X-Sylpheed-Privacy-System:
X-Sylpheed-Sign:0
SCF:#mh/Mailbox/sent
RMID:#imap/andys@onstor.net@onstor-exch02.onstor.net/INBOX	0	BB375AF679D4A34E9CA8DFA650E2B04E056C9105@onstor-exch02.onstor.net
X-Sylpheed-End-Special-Headers: 1
Date: Mon, 24 Sep 2007 16:25:20 -0700
From: Andrew Sharp <andy.sharp@onstor.com>
To: "Larry Scheer" <larry.scheer@onstor.com>
Subject: Re: review of emrscron, autosupport changes
Message-ID: <20070924162520.4e3ad25a@ripper.onstor.net>
In-Reply-To: <BB375AF679D4A34E9CA8DFA650E2B04E056C9105@onstor-exch02.onstor.net>
References: <20070924150043.0af0e5f6@ripper.onstor.net>
	<BB375AF679D4A34E9CA8DFA650E2B04E056C9105@onstor-exch02.onstor.net>
Organization: Onstor
X-Mailer: Sylpheed-Claws 2.6.0 (GTK+ 2.8.20; x86_64-pc-linux-gnu)
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

Here is what my script does:

P4C_SPEC=`p4 describe $CLN | grep "Change $CLN" | sed 's/.*@\([^ ]*\).*/\1/'`

...

p4 describe $CLN | while read DOTS YICK ACTION GARBAGEE ; do

   ...

   P4F=`echo $YICK | sed "s+#.*$++"`

   ...

   wsfile=`p4 -c "$P4C_SPEC" have "$P4F" | cut -d' ' -f 3`



As far as I can tell, that should work.  It works on some of the files
in the cl.



On Mon, 24 Sep 2007 15:13:38 -0700 "Larry Scheer"
<larry.scheer@onstor.com> wrote:

> Rendell's workspace for his dev branch is /homes/rendellf/dev not
> /homes/rendellf/p4dev.
> 
> What I do to figure out the client's workspace when I am not given it
> in the email is I do a p4 describe change# and look for the client
> name which will be specified as user@client in this case it is
> rendellf. Then I do a p4 client -o client-name, again rendellf in
> this case, look at what Root is set to (/ for this Rendell's client)
> and then the branch specification
> (//depot/dev/... //rendellf/homes/rendellf/dev/...) which tells me
> his workspace is /homes/rendellf/dev.
> 
> Happy hunting,
> 
> Larry
> 
> -----Original Message-----
> From: Andy Sharp 
> Sent: Monday, September 24, 2007 3:01 PM
> To: Larry Scheer
> Cc: Rendell Fong
> Subject: Re: review of emrscron, autosupport changes
> 
> OK, I tried to review more of the files, see my comments at the end.
> Many of the files are not available, I presume they aren't really part
> of this cl?  Checked out on a different client?  Major bug in my
> review script?  Some kind of permissions problem in some
> directories??  Hmm, I don't get it, they are listed in the
> changelist, but still perforce is giving me that error on some of the
> files.  Damnit.
> 
> On Mon, 24 Sep 2007 14:07:35 -0700 "Larry Scheer"
> <larry.scheer@onstor.com> wrote:
> 
> > Rendell,
> >    Emrscron line 324 just a style change... - well there are many I
> > should suggest but here is one that I think should become  a habit.
> > When combining shell variables with string constants use curly
> > braces. Eample Instead of mount -$OPT"w" use mount -${OPT}w. Also
> > you should do this when combining variables:
> > mount -$OPT$TYPE should be mount -${OPT}${TYPE} 
> > 
> > Another thing I seen with support.sh is stuff like if [ $OS ==
> > "Linux" ] you should get in the habit of  always quoting the
> > variable. Then the shell won't die if the variable is empty. Of
> > course then if it is empty when you quote it and it is supposed to
> > have a value the your script is running with a subtle, hard to find
> > bug, but what the heck at least you have "good" coding style! ;-)
> > Actually there is a shell option that you can turn on to catch
> > un-initialized variables. 
> > 
> > Only skimmed the c source changes you made so far I thought they
> > were ok to check in since I don't know a whole lot about what the
> > changes should be. I am also hoping these changes may change the
> > behavior of the eventd crash I am seeing. Maybe Andy has looked at
> > them closer.
> > 
> > Larry
> > _____________________________________________
> > From: Rendell Fong 
> > Sent: Wednesday, September 19, 2007 12:50 PM
> > To: Larry Scheer
> > Cc: Andy Sharp
> > Subject: review of emrscron, autosupport changes
> > 
> > Larry,
> > 
> > Assuming the initial config changes are working, I believe that all
> > the code changes are completed.
> > Please review these numerous changes.
> > 
> > 
> > Andy,
> > 
> > I've made changes to the initial-config file to setup exim config
> > files with the hostname specified by the user.
> > You may want to review it. 
> > 
> > Thanks guys,
> > Rendell
> > 
> > --------------------------------------------
> > 
> > Change #25481
> > 
> > Description:
> >     TED20323 - Fixed autosupport and emrs for Linux.  Moved
> > processing of
> >     autosupport mail server from CLI autosupport cmd to autosupport
> > daemon.
> >     Sync exim/sendmail config files with cluster DB config upon
> > autosupport
> >     daemon initialization.  Set hostname in exim config files during
> > initial
> >     config script.
> > 
> > Files:
> >     //depot/dev/linux/rootfs/etc/exim4/update-exim4.conf.conf   #
> > add //depot/dev/linux/rootfs/etc/onstor/initial-config  # edit
> >     //depot/dev/nfx-tree/code/sm-event/Makefile # edit
> >     //depot/dev/nfx-tree/code/sm-event/asd.c    # edit
> >     //depot/dev/nfx-tree/code/sm-event/event-as-linux.c # add
> >     //depot/dev/nfx-tree/code/sm-event/event-as-openbsd.c   # add
> >     //depot/dev/nfx-tree/code/sm-event/event-as.c   # edit
> >     //depot/dev/nfx-tree/code/sm-event/linux.h  # add
> >     //depot/dev/nfx-tree/code/sm-event/openbsd.h    # add
> >     //depot/dev/nfx-tree/code/ssc-nfxsh/cmd_autosupport.c   # edit
> >     //depot/dev/nfx-tree/code/ssc-support/emrscron  # edit
> >     //depot/dev/nfx-tree/code/ssc-support/support.sh    # edit
> > 
> 
> 
> 
> linux/rootfs/etc/exim4/update-exim4.conf.conf
> 
>      actually, the dc_other_hostnames should be set to '' and does not
>      need to be set to the hostname of the filer.
>      setting the /etc/mailname file should be all that needs doing,
>      so remove the engXX.
> 
>      IIRC dc_other_hostnames is for host names that we would allow
>      relaying for, which is none.
> 
> linux/rootfs/etc/onstor/initial-config
> 
>      lines 1626-1635 should be removed.  exim should still be
> restarted, however, as /etc/mailname has changed.  And that makes all
> the difference.
> 
> nfx-tree/code/sm-event/Makefile
> 
>      i get "/homes/rendellf/p4dev/nfx-tree/code/sm-event/Makefile -
>      file(s) not opened on this client."
> 
> nfx-tree/code/sm-event/asd.c
> 
>      i get "/homes/rendellf/p4dev/nfx-tree/code/sm-event/asd.c -
>      file(s) not opened on this client."
> 
> nfx-tree/code/sm-event/event-as-linux.c
> 
>      >>add nfx-tree/code/sm-event/event-as-linux.c
> 
>      please remove trailing whitespace, lines 18-21, 75, 152, 165, 245
> 
>      line 113 do no put else on a line by itself after a right curly;
>      make it 'else if' together on the same line (matching style of
>      line 123):
> 
>      } else if (chewbaca == ape) {
> 
> 
>      All these EXIM_MAILMODE_TOK macros, why are they
>      defined somewhere else and not in this file?  Are they used in
>      other files?  Define them here, where they are used.
> 
>      line 184 (and others)
> 
>      you don't need to cast these calls to void, so please don't.
>      you don't on 148. and please use the remove(3) function instead
>      of unlink(2).
> 
>      line 238, as_syncMailAgt()
> 
>      this is not the way to do this.  read /var/run/exim4.pid and
>      sighup that pid, or just do a system("/etc/init.d/exim4
> restart").
> 
> nfx-tree/code/sm-event/event-as-openbsd.c
> 
>      >>add nfx-tree/code/sm-event/event-as-openbsd.c
> 
>      again, remove trailing whitespace please.  lines 18-21, 75, etc.
> 
>      since openbsd is essentially dead, i don't care about the rest
>      of this ~:^)
> 
> nfx-tree/code/sm-event/event-as.c
> 
>      i get "/homes/rendellf/p4dev/nfx-tree/code/sm-event/event-as.c -
>      file(s) not opened on this client."
> 
> nfx-tree/code/sm-event/event-as.h
> 
>      i get "/homes/rendellf/p4dev/nfx-tree/code/sm-event/event-as.h -
>      file(s) not opened on this client."
> 
> nfx-tree/code/sm-event/linux.h
> 
>      >>add nfx-tree/code/sm-event/linux.h
> 
>      uh, i'd rather these defines were in the C file that uses them,
>      unless they are used in more than one.  at least the strink
>      token macros.
> 
> nfx-tree/code/sm-event/openbsd.h
> 
>      >>add nfx-tree/code/sm-event/openbsd.h
> 
>      ok, looks good.
> 
> nfx-tree/code/ssc-nfxsh/cmd_autosupport.c
> 
>      I get
>      "/homes/rendellf/p4dev/nfx-tree/code/ssc-nfxsh/cmd_autosupport.c
> - file(s) not opened on this client."
> 
> nfx-tree/code/ssc-support/emrscron
> 
>      i get "/homes/rendellf/p4dev/nfx-tree/code/ssc-support/emrscron -
>      file(s) not opened on this client."
> 
> nfx-tree/code/ssc-support/support.sh
> 
>      i get
> "/homes/rendellf/p4dev/nfx-tree/code/ssc-support/support.sh - file(s)
> not opened on this client."
> 
> Perfarce is really making me mad with these errors.
> 
