[OpenAFS] Speed of release or incremental release (replication)?

[Matthew E Hoskins - SAGE AFS]

Ive been wathing this thread for a little while, I just wanted to point
out one thing...  When AFS releases a vol it tries to make the lock on
the RW vol as short as possible.  It does this by releasing to a RO on
the local server first.  This is why the docs state that there should be
an RO on the same part as the RW.  When none exists, or the RO is on a
dif part, afs creates a temp release clone.   This can take a very long
time.  

> I'd like to have the most important volumes replicated, but as it
> is today, even twice a day is to much hassle (it locks up during
> release).
> 
> ----- s n i p -----
> Releasing volume user.turbo.mail (20031023 12:00:49):
> Cloning RW volume to temporary RO ...                     <------ See this
> 
> user.turbo.mail
>     RWrite: 536871355     ROnly: 536871356
>     number of sites -> 2
>        server papadoc.bayour.com partition /vicepc RW Site   <--- Because this
>        server papadoc.bayour.com partition /vicepe RO Site   <--- And this
> This is a complete release of the volume 536871355
> Updating existing ro volume 536871356 on papadoc.bayour.com ...
> Starting ForwardMulti from 536871356 to 536871356 on papadoc.bayour.com.
> Deleting the releaseClone 536871356 ... done
> updating VLDB ... done
> Released volume user.turbo.mail successfully
> Volume released at 20031023 14:02:31. Took 7302 sec (~121 min)
> This means that 146 Kb was released per second
> ----- s n i p -----
> 
> It say 'This is a complete release of the volume' so this leads
> me to think that it should be possible to do a 'non-complete'
> (ie, incremental) release... ?
> 
> And 146Kb/sec seems HORRIBLE to me! This is from one partition
> (ie, disk) to another. No 'network connection' is done. I do
> know that AFS in it self is through the network layer so some
> overhead is expected, but that much!?
> 
> I MUST have configured something wrong, because I can't believe
> that AFS is this slow!!
> 
> The machine is a dual PII 333MHz with Ultra160 disks 'in full
> working order'... The machine IS loaded, but not that much...
> 
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info

-- 
# Matthew E. Hoskins  #############################################
# Unix System Administrator              _/|  _/     _/ _/  __/__ #
# University Information Systems        _/_| _/     _/ _/   _/    #
# New Jersey Institute of Technology   _/ _|_/ _/  _/ _/   _/     #
# University Heights, Newark, NJ 07102_/  _|/ _/___/ _/   _/      #
# Ph: 973-596-5202         #  Beep: pagematt@cobol.njit.edu       #
# Rm: 5400 GITC Building   # Email: matt@njit.edu                 #
###################################################################
"Any technology sufficiently advanced is indistinguishable 
        from a perl script" -Unknown
"Anyone who considers arithmetical methods of producing 
        random digits is, of course, in a state of sin"
	-- John von Neumann