RE: What IOS really does with commands (was Re: ISIS Problem. posted 03/31/2004
Something for you all to note is that many routers and IOS versions when you do a wr mem you induce a 350ms
Outage of traffic flowing through the box. Now some of you may say "big deal" but put a voice switch or a time
Sensitive application and watch it die every time you save the config. Some IOS versions now only do this the first
Time you save the config i.e. just after it is reloaded others still do it every time. Also changing the boot system commands induces the same delay (there is no way round this).
CCIE #8757 R&S
> -----Original Message-----
> From: dburnett@xxxxxxxxxx [mailto:dburnett@xxxxxxxxxx]
> Sent: 31 March 2004 16:25
> To: Howard C. Berkowitz
> Cc: ccielab@xxxxxxxxxxxxxx
> Subject: Re: What IOS really does with commands (was Re: ISIS Problem.
> Where may someone find more information (as much as may be publicly
> available) about how IOS creates and stores the data
> structure that it modifies from entered keywords (aka commands)?
> ----- Original Message -----
> From: "Howard C. Berkowitz" <hcb@xxxxxxxxxxxx>
> To: <ccielab@xxxxxxxxxxxxxx>
> Sent: Sunday, March 28, 2004 9:20 PM
> Subject: What IOS really does with commands (was Re: ISIS Problem.
> > At 7:56 AM -0800 3/28/04, Joseph Rinehart wrote:
> > >?Sometimes the router will slightly alter commands in the active
> > >configuration when you enter them. I had this happen on an
> > >access-list
> > >week and it nearly drove me crazy at first.
> > Something worth knowing to anyone studying for CCIE, but rarely
> > spelled out specifically, is that the router doesn't just
> record, or
> > even alter the commands when you enter them -- it
> interprets them and
> > then throws them away. The earlier courses can give you the
> > impression that the commands are entered are stored
> somewhere in RAM.
> > They are not.
> > On slower routers, did you ever notice the slight delay
> before "show
> > running" displays? Show running is essentially a reverse
> compiler or
> > formatted dump that goes throgh the data structures of the
> router and
> > creates the configuration statements it thinks would have put the
> > router into the states reflected by the current values of the data
> > structures. The delay is due to the significant processing
> > in recreating the statements.
> > >
> > >I think this is what is happening, and you would need to
> change it by
> > >specifying Level-1 in the redistribution statement, I believe the
> > >syntax
> > >"redistribute isis 1 level-1-only metric 4"
> > _
> > Please help support GroupStudy by purchasing your study
> materials from:
> > http://shop.groupstudy.com
> > Subscription information may be found at:
> > http://www.groupstudy.com/list/CCIELab.html
> Please help support GroupStudy by purchasing your study
> materials from: http://shop.groupstudy.com
> Subscription information may be found at:
Please help support GroupStudy by purchasing your study materials from:
Subscription information may be found at: