AF:
NF:0
PS:10
SRH:1
SFN:
DSR:
MID:<20080403133519.7e8018a8@ripper.onstor.net>
CFG:
PT:0
S:andy.sharp@onstor.com
RQ:
SSV:onstor-exch02.onstor.net
NSV:
SSH:
R:<sripal.surendiran@onstor.com>,<john.keiffer@onstor.com>,<larry.scheer@onstor.com>,<sandrine.boulanger@onstor.com>,<vikas.saini@onstor.com>,<tim.gardner@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	47F52F3D.5020004@onstor.com
X-Sylpheed-End-Special-Headers: 1
Date: Thu, 3 Apr 2008 13:36:53 -0700
From: Andrew Sharp <andy.sharp@onstor.com>
To: Sripal <sripal.surendiran@onstor.com>
Cc: John Keiffer <john.keiffer@onstor.com>, Larry Scheer
 <larry.scheer@onstor.com>, Sandrine Boulanger
 <sandrine.boulanger@onstor.com>, Vikas Saini <vikas.saini@onstor.com>, Tim
 Gardner <tim.gardner@onstor.com>
Subject: Re: Defect  TED00022980 OCT / FTI - I answered "yes" in the OCT
 screen to copy the config files, then it fails
Message-ID: <20080403133653.14468da4@ripper.onstor.net>
In-Reply-To: <47F52F3D.5020004@onstor.com>
References: <BB375AF679D4A34E9CA8DFA650E2B04E0941A882@onstor-exch02.onstor.net>
	<20080403114618.317088ff@ripper.onstor.net>
	<47F52F3D.5020004@onstor.com>
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


In this isolated case, the configuration file didn't have a valid
format because the format changed.  Normally that wouldn't happen
except that we are developing this Linux based distribution for the
first time.  In a normal case, the upgrade software will take care of
the task of migrating the file to the new format.

In the rare case a config file is corrupted or missing for some other,
unexpected reason, a meaningful error message is all that we can do.
We should not copy the config files in a case such as that.

Cheers,

a

On Fri, 04 Apr 2008 00:55:49 +0530 Sripal
<sripal.surendiran@onstor.com> wrote:

> John,
> 
> Load from secondary option will fail only if the secondary flash
> doesn't have valid mandatory configuration file. This is how it
> worked even in Rocket98 for BSD (Bug TED00021140 has been filed to
> check the valid configuration file in secondary before loading the
> files). Hence I am insisting on WAD.
> 
> You will not face the problem if you do flash install. If we accept
> to not care for valid files in secondary, we may face issues like
> loading configuration files in secondary flash that is in
> uninitialized state. So during the next boot up we may see filer not
> having valid/desired IP address. User has to connect only using
> console port.
> 
> 
> Hope this answers your questions.
> 
> -Sripal.
> 
> Andrew Sharp wrote:
> > This is a case where config file format changes have occurred, so if
> > you upgrade to software that requires those changes but your config
> > files are still using the old format, this condition will occur.  We
> > can debate about whether it's a bug or not, but we won't fix it.  In
> > this case, the files will have to be converted by hand as described.
> > As I mentioned to Vikas, after we have a shipping release, these
> > kind of changes will of course be handled by the upgrade software,
> > but not before then, because doing so would cause unacceptable
> > bloat.  The idea being that at some point prior to shipping the
> > first version, the frequency of such changes will diminish to the
> > point that putting such code in the upgrade software won't bloat
> > the the software unnecessarily.
> >
> > Feel free to discuss more if you have more questions.
> >
> > Cheers,
> >
> > a
> >
> > PS Software Dev failed in it's duty to properly inform QA of the
> > implications of this change.  For that, we apologize and will try
> > to do a better job in the future.
> >
> >
> > On Thu, 3 Apr 2008 11:28:38 -0700 "John Keiffer"
> > <john.keiffer@onstor.com> wrote:
> >
> >   
> >> Gentlemen,
> >>
> >> Your expertise in all things with boot up, install, and upgrades is
> >> needed on helping us with understanding this defect. Sripal keeps
> >> wanting to close this defect as a WAD, but I am not able to
> >> understand the exact reason he thinks it might be okay to have a
> >> secondary flash config copy fail in the new FTI. 
> >>
> >> What's to stop a customer from taking a flash card and trying to
> >> copy the config from it while in the FTI? I mean... How can we be
> >> clear when the config copy will work and when it won't? Will we
> >> make it a requirement that the user understand and edit the files
> >> Sripal mentions in the defect, before they try this process in the
> >> FTI?
> >>
> >> Any clarifications you can provide would be greatly appreciated.
> >>
> >> Thanks,
> >> John
> >>
> >> -----Original Message-----
> >> From: Sandrine Boulanger 
> >> Sent: Thursday, April 03, 2008 10:19 AM
> >> To: John Keiffer
> >> Cc: Vikas Saini
> >> Subject: RE: Defect TED00022980 OCT / FTI - I answered "yes" in the
> >> OCT screen to copy the config files, then it fails
> >>
> >> We should discuss this with Larry and Andy, I still don't think
> >> this is WAD.
> >>
> >> -----Original Message-----
> >> From: John Keiffer 
> >> Sent: Thursday, April 03, 2008 7:44 AM
> >> To: Sandrine Boulanger
> >> Cc: Vikas Saini
> >> Subject: FW: Defect TED00022980 OCT / FTI - I answered "yes" in the
> >> OCT screen to copy the config files, then it fails
> >>
> >>
> >> I'm getting confused by this. Shouldn't this just plain old work?
> >>
> >> -----Original Message-----
> >> From: sripal.surendiran@onstor.com
> >> [mailto:sripal.surendiran@onstor.com]
> >>
> >> Sent: Thursday, April 03, 2008 7:11 AM
> >> To: Sripal Surendiran (HCL); Tim Gardner; John Keiffer
> >> Cc: John Keiffer; Vikas Saini
> >> Subject: Defect TED00022980 OCT / FTI - I answered "yes" in the OCT
> >> screen to copy the config files, then it fails
> >>
> >> Headline: OCT / FTI - I answered "yes" in the OCT screen to copy
> >> the config files, then it fails
> >> id: TED00022980
> >> Note_Entry: It is not that both eth0 and eth1 entry should be
> >> configured.  Only entry for both the interface should be present in
> >> the file. For those interface that are not configured, the entry
> >> should be commented out. Even if FTI ignores the format, we may
> >> face problems in ipmd. Please let us know your thoughts.
> >> Otherwise, I am planning to close this bug again as WAD.
> >> State: Reviewed
> >> history: 33745368	Mar 26 2008  7:32AM	jkeiffer
> >> Submit no_value	Opened
> >> 33745372	Mar 26 2008  8:38AM	jkeiffer
> >> Modify	Opened Opened
> >> 33745374	Mar 26 2008  9:01AM	jkeiffer
> >> Modify	Opened Opened
> >> 33745426	Mar 26 2008  1:49PM	timg	Assign
> >> Opened	Assigned 33745432	Mar 26 2008  1:59PM
> >> jkeiffer	Modify	Assigned Assigned
> >> 33745450	Mar 26 2008  3:06PM	sandrineb
> >> Modify	Assigned Assigned
> >> 33745547	Mar 27 2008  8:44AM	sripals	WAD
> >> Assigned WAD
> >> 33746193	Apr  1 2008  1:21PM	jkeiffer
> >> Close	WAD Closed
> >> 33746305	Apr  2 2008 10:36AM	jkeiffer
> >> Limbo	Closed Limbo
> >> 33746306	Apr  2 2008 10:36AM	jkeiffer
> >> Back_to_triage Limbo	Re-opened
> >> 33746437	Apr  2 2008  3:47PM	vikas	Modify
> >> Re-opened Re-opened
> >> 33746438	Apr  2 2008  3:47PM	vikas	Review
> >> Re-opened Reviewed
> >> 33746511	04/03/2008 07:11:05 AM	sripals
> >> Modify	Reviewed Reviewed
> >>
> >>     
> 
