LSTSRV-L Archives

LISTSERV Site Administrators' Forum

LSTSRV-L

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Topic: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
"A. Harry Williams" <[log in to unmask]>
Tue, 9 Dec 2008 12:37:21 EST
TEXT/PLAIN (52 lines)
On Tue, 9 Dec 2008 10:52:06 -0500  said:
>David Young wrote on 12/09/2008 10:33:19 AM:
>
>
>> The individual command already uses the * wildcard character, such as:
>>
>> Quiet change * [log in to unmask] [log in to unmask]
>>
>> If the usernames remained the same, it wouldn't be a problem, but
>> they're more personalized on the new e-mail host.  Fo instance:
>>
>> Quiet change * [log in to unmask] [log in to unmask]
>>
>> I hope that helps explain things.
>
>>> ====
>>> CHANGE(*) listname|* newaddr
>>> CHANGE(*) listname|* oldaddr|pattern newaddr|*@newhost
>>> ====
>
>If the LHS was not changing you could use the second form of the command:
>
>QUIET CHANGE * *@exchange.uta.edu *@mavs.uta.edu
>
>Looks like you're going to have to script the changes since the LHS is
>changing too.  Maybe you can have all of your script files written with 100
>changes each.  Then time how long it takes to process each one and write a
>shell script to launch each set of change commands at intervals of 2 -4
>times (or more) the time it take to process each set.  That way, you would
>only batter and bruise, not completely crush  the server.


I would also look and see if you can use the list-specific format.

Quiet change list1 [log in to unmask] [log in to unmask]

is significantly faster than

Quiet change * [log in to unmask] [log in to unmask]

even if you have to do it 3 or 4 times if you have a significant number
of lists.  Even if listserv changes the files while you are processing
and your changes are only 90% of the people/list combination, it will be
significantly faster.  since it appears that both forms of addresses will
be valid for a period of time, you could also rerun the process after
converting to catch those that you missed the first time an potentially
leave only 1% to be done by the longer methods)  This will also be true if
some substantial portion of the 10,000 users are not subscribed to any of
the lists you are looking at. (If you use the wildcarded list method, you
also will search and potentially update those lists you mentioned that are
automatically rebuilt)

ATOM RSS1 RSS2