Networker

Re: [Networker] NWRestore taking FOREVER to complete???

2003-08-01 11:56:40
Subject: Re: [Networker] NWRestore taking FOREVER to complete???
From: "Maarten Boot (CWEU-USERS/CWNL)" <maarten.boot AT NL.COMPUWARE DOT COM>
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Date: Fri, 1 Aug 2003 17:56:35 +0200
Dont know anything about Netware,

Restoring always causes surprises the first time you encounter them,
speciaclly large amounts.

Note that several crucial aspects play a role here,

data may be multiplexed on tape that can cause a delay and if your data is
lots of small files that can cause a major delay aswell as your OS and the
disks are writing may small files instead of several large data streams.

I recently had that experience when we had to restore several hundered
gigabytes. the home directories of users were 4 times slower thamn the backup
partition which hold backups of our machines bihind the firewall. those files
tend to be a few very large files of 2 Gb or more.

So lots of factors are involved and anyway writing to disk (specially raid 5)
is in general much slower than reading so backup times say nothing about
restore times. (my rule of thumbe is 4 times slower)

Maarten


On Friday 01 August 2003 16:17, you wrote:
> ENVIRONMENT:
>
> STORAGE NODE:
> Legato Neworker 6.1.3 running on a dual Xeon processor SuperMicro server,
> RedHat 7.3
> 100 mb NIC - Full duplex
> dual Overland LibraryPro (Sony 700 AIT-3) jukeboxes
> small block sizes, as we are not using an STINIT.DEF file just yet.  Plan
> to get that implemented as soon as life returns to normal when this
> server's restored.  Block size is 512
>
> Restoring:
> Novell Netware 4.12 running on a single processor HP LPr Server,
> with a Cybernetics SCSI RAID array -
> 100 mb NIC - Full duplex
> Using Networker 4.21 client
>
> TWO QUESTIONS:
>
> 1st Q - The Novell server crashed and burned two weeks ago - lost a
> RAID.  We have three volumes, total capacity of 51 gigs in three volumes,
> but not all that space is used, that's what we're restoring.  The restore
> began yesterday at about 2:30 p.m. (some 19 hours later), and is still
> running!  My NWAdmin screen shows I've only restored 6.7 of 15 Gigs!!!
>
> There's data going to the new server, we can watch files and directories
> being restored, with the proper ownership rights...  We restored from data
> that was backed up to a laptop something over 3 gigs of space, which took
> 15 minutes max to copy.
>
> WHY is it sooooooo slow?  Is this related to the block size, slowing down
> the tape access/read?
>
> 2nd Q - Our Novell consultant advised the server abended recently due to
> what they determined to be a memory leak in the Networker Netware 4.21
> client we're using.  They recommend we reboot the Novell server quarterly
> to avoid the problem.
>
> HAS ANYONE else heard of a memory leak in this Networker Netware 4.21
> client?
>
> Thank you in advance,
>
> Les Yaw
> Luther College
> Decorah, IA
>
>
> --
> Note: To sign off this list, send a "signoff networker" command via email
> to listserv AT listmail.temple DOT edu or visit the list's Web site at
> http://listmail.temple.edu/archives/networker.html where you can
> also view and post messages to the list.
> =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

<Prev in Thread] Current Thread [Next in Thread>