AF:
NF:0
PS:10
SRH:1
SFN:
DSR:
MID:
CFG:
PT:0
S:andy.sharp@lsi.com
RQ:
SSV:mhbs.lsil.com
NSV:
SSH:
R:<Larry.Scheer@lsi.com>,<Rendell.Fong@lsi.com>,<Yifeng.Liu@lsi.com>,<Elizabeth.George@lsi.com>,<dl-engineering@lsi.com>,<Arindam.Banerjee@lsi.com>,<Aaron.Dailey@lsi.com>,<Brian.Stark@lsi.com>
MAID:2
X-Sylpheed-Privacy-System:
X-Sylpheed-Sign:0
SCF:#mh/Mailbox/sent
RMID:#imap/LSI/INBOX	0	DEC609CD0E54B2448DAF023C89AE9755E92759C5@cosmail02.lsi.com
X-Sylpheed-End-Special-Headers: 1
Date: Mon, 12 Oct 2009 12:12:25 -0700
From: Andrew Sharp <andy.sharp@lsi.com>
To: "Scheer, Larry" <Larry.Scheer@lsi.com>
Cc: "Fong, Rendell" <Rendell.Fong@lsi.com>, "Liu, Yifeng"
 <Yifeng.Liu@lsi.com>, "George, Elizabeth" <Elizabeth.George@lsi.com>,
 DL-ONStor-Engineering <dl-engineering@lsi.com>, "Banerjee, Arindam"
 <Arindam.Banerjee@lsi.com>, "Dailey, Aaron" <Aaron.Dailey@lsi.com>, "Stark,
 Brian" <Brian.Stark@lsi.com>
Subject: Re: Question on ONStor NAS product logging
Message-ID: <20091012121225.7cbe6a1c@ripper.onstor.net>
In-Reply-To: <DEC609CD0E54B2448DAF023C89AE9755E92759C5@cosmail02.lsi.com>
References: <D36DB4126B506541A862C33BE1FB1488015F3FEBA2@cosmail03.lsi.com>
	<861DA0537719934884B3D30A2666FECC94455D8C@cosmail02.lsi.com>
	<EAD7AABE04CB5041A411B73451B8F0880405CB8C@cosmail03.lsi.com>
	<1255371794.14287.317.camel@rendellf>
	<DEC609CD0E54B2448DAF023C89AE9755E92759C5@cosmail02.lsi.com>
Organization: LSI
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=UTF-8
Content-Transfer-Encoding: quoted-printable

On Mon, 12 Oct 2009 12:37:23 -0600 "Scheer, Larry"
<Larry.Scheer@lsi.com> wrote:

> One piece of information requested that was missing from previous
> responses...
>=20
> The size of the /var partition on the compact flash is: 16.8 Mbytes.
> This is where all the system logs, elogs, trace files, crash files
> and user applications and daemon core files are stored.

Slight correction: the /var partition is closer to 170MiB.  Just off by
a silly little zero ~:^)

> Larry
> ________________________________________
> From: Rendell Fong [Rendell.Fong@lsi.com]
> Sent: Monday, October 12, 2009 11:23 AM
> To: Liu, Yifeng
> Cc: George, Elizabeth; DL-ONStor-Engineering; Banerjee, Arindam;
> Dailey, Aaron; Stark, Brian Subject: RE: Question on ONStor NAS
> product logging
>=20
> Comments are inline.
>=20
> On Mon, 2009-10-12 at 09:11 -0600, Liu, Yifeng wrote:
> > Hi all,
> >
> >
> >
> > Liz is working on centralized logging in the Pikes Peak project and
> > has few questions regarding various areas of ONStor NAS GW, would
> > you please help her in finding answers, thanks.
> >
> >
> >
> > Yifeng
> >
> >
> >
> > From: George, Elizabeth
> > Sent: 2009=E5=B9=B410=E6=9C=889=E6=97=A5 15:32
> > To: Liu, Yifeng
> > Cc: George, Elizabeth; Banerjee, Arindam; Dailey, Aaron; Stark,
> > Brian Subject: Question on ONStor NAS product logging
> >
> >
> >
> >
> > Yifeng,
> >
> > Thanks for your help so far.. Some more questions coming your way on
> > current OnStor NAS product.  Please feel to forward it as needed, if
> > it is not in your area of expertise.
> >
> >      1. What is current size and persistent storage location of
> > elogs?
>=20
> The elogs are stored in /var/log/onstor.  Message files are logrotated
> periodically (daily) and perhaps more often based on max size (1M
> bytes) as needed.
>=20
> >      1. What is current size and persistent storage location of
> >         corefiles (the trace info)?
>=20
> For user space daemons (apps), core files are stored in /var/run.
> Linux generates the core file and the size will vary based on the
> crash and app.  Not sure what the typical core file size is.  Maybe
> someone else knows.
>=20
> >      1. How is the system log  different from Corefiles? What is
> > it=E2=80=99s size? And where is it stored?
>=20
> The system log is stored in /var/log and records syslog type events
> posted by the kernel and TXRX/FP code.  The system log files are
> logrotated daily.
>=20
> >      1. What is current size and storage location of coredump? Any
> >         compression algorithm/checksumming algorithm used?
>=20
> For TXRX and FP, coredump files are stored in the mgmt volume are
> typically 25-50M bytes, gzipped.  The coredump file is stored in ELF
> format.  No checksumming is done on coredump file data.
>=20
> Crashdump files are stored in /var/crash.  A crashdump file entry is
> generated for each TXRX and FP crash and appended to the file if it
> already exists.  The entry includes a register set snapshot and call
> backtrace (w/o symbols) in text format.  When crash info is
> temporarily stored in prom at time of crash, a checksum is computed
> for it.  The checksum is a simple addition of data byte values.  No
> checksum is used for the info once it is copied to the crashdump file.
>=20
> >      1. Can current elog identify , on each event, which node it was
> >         logged by?
>=20
> elog events do show the node name and entity that logged it.
>=20
> Example:
> Oct 12 08:22:28 eng59 : 1:3:coredump:NOTICE: 1866: core_file_thread:
> Sleeping for 1 hour and try copying core dump later
>=20
> In this case, eng59 is the node and coredump is the entity.
>=20
> >      1. What if any, is the level of criticality indication inside
> > an event, other than the flag to determine if ASD should send
> >         email or not?
> >
> > Ie Would you distinguish between severity of an event?
> >
>=20
> Severity level is not being used to determine if email should be sent
> or not.  Sending of autosupport email can only be enabled or
> disabled.  The event type determines whether an email is generated.
> This determination cannot be selectively changed by the user.
>=20
> >
> >
> > Thanks so much with your patience in answering my numerous
> > questions.
> >
> > -Liz
> >
> >
>=20
