I'm not sure how Networker handles timezones. I know I've discovered
clients with out of sync clocks when they were missing from my clone script
which used savetime. I then changed it to sscreate which referenced the time
on the server. Specifying to clone data with a savetime greater than 23:00
missed a saveset whose savetime is 22:59; the saveset had a savetime
that was prior to the group starting. Use of a ntp or other time server
is wise...one may not have control over the time on a client though.
It my understanding that it does not matter whether you run the commands
on the client or server..they all query networker on the server.
Robert Maiello
Pioneer Data Systems
On Wed, 20 Oct 2004 12:00:00 -0400, George Sinclair
<George.Sinclair AT NOAA DOT GOV> wrote:
>Yeah, would be nice to know which it is. I think I'd rather use the time
>created on the server and not the time from the client. That just seems
>more official. By client, do you mean the client I'm running the mminfo
>command from or the time the saveset was created on the respective
>client?
>
>I assume if I do want to use the time created on the server then it
>won't matter that I'm running the mminfo command from a client, correct?
>As long as the time I'm specifying, e.g. -t '3 days ago 4 pm' is a time
>used on the server. In other words, server is set to EST or EDT
>depending, and so is client, but if client was set to UTC then would
>recieve different output.
>
>George
>
>
>
>
>Robert Maiello wrote:
>>
>> The mminfo manpage is ambiguous as to what "savetime" the -t refers to.
>> Is it the same as the savetime attribute (which is the time on the client)
>> or is it the sscreate attribute (which is the time it was created on
>> the server) ??
>>
>> If one of your clients has a clock a bit out sync it may go missing from
>> reports?
>>
>> Robert Maiello
>> Pioneer Data Systems
>>
>> On Wed, 20 Oct 2004 11:32:34 -0400, George Sinclair
>> <George.Sinclair AT NOAA DOT GOV> wrote:
>>
>> >Well, what I've discovered is that even with the '-a' option, using '-t'
>> >does NOT list index entries, bootstrap or any savesets wherein the pool
>> >has indexing turned off. However, indexes, bootstrap and savesets from
>> >pools with indexing turned off *DO* show up when using the 'saveset'
>> >attribute. I then tried using '-v' option with the '-a' and '-t' as:
>> >
>> >mminfo -s server -av -t 'yesterday 4 pm' or 'yesterday 16:00'
>> >
>> >and then I see the index, bootstrap and other entries that were missing
>> >when run without the 'v'. Of course, to get the output to match the
>> >output from the mminfo command that uses the saveset attribute, I
>> >likewise have to include '-v' option on that. What the heck is going on
>> >here?!!!! This seems ridiculous.
>> >
>> >Another thing I notice is that if I run the command with the saveset
>> >attribute, but I grep out the index, bootstrap and the one saveset name
>> >(special) whose pool has indexing turned off as:
>> >
>> >mminfo -s server -q 'savetime>=yesterday 4pm' | grep -v "index:" | grep
>> >-v bootstrap | grep -v
>> >
>> >and then I compare with:
>> >
>> >mminfo -s server -t 'yesterday 4pm'
>> >
>> >then it matches, but if I go back a couple of days, I find a few
>> >savesets that still show up in the first command and not the second, but
>> >only a couple, and the pools that these write to do not have indexing
>> >turned off. Based on what I've found, I just don't trust the '-t' option
>> >to really give you everything. I have a sneaking suspicion that it will
>> >leave something out that you were expecting.
>> >
>> >George
>> >
>> >
>> >Darren Dunham wrote:
>> >>
>> >> >
>> >> > Hi,
>> >> >
>> >> > I notice I receive identical output for all of the following commands
>> >> > wherein I list savesets since 4 PM (EDT) yesterday:
>> >> >
>> >> > mminfo -s server -q 'savetime>=10/18/04 4 pm'
>> >> > mminfo -s server -q 'savetime>=10/18/04 16:00'
>> >> > mminfo -s server -q 'savetime>=1098129600'
>> >>
>> >> This uses -q, which implies -a.
>> >>
>> >> > HOWEVER, if I instead run:
>> >> >
>> >> > mminfo -s server -t 'yesterday 4 pm' or 'yesterday 16:00'
>> >>
>> >> This doesn't use q, c, m, N, or o and you haven't explicitly used -a.
>> >>
>> >> My assumption is that this is the difference. Because I've been bitten
>> >> by similar issues so many times, I just always use -a.
>> >>
>> >> See if that makes everything work the same.
>> >>
>> >> --
>> >> Darren Dunham ddunham AT taos
>> >> DOT com
>> >> Senior Technical Consultant TAOS http://www.taos.com/
>> >> Got some Dr Pepper? San Francisco, CA bay area
>> >> < This line left intentionally blank to confuse you. >
>> >>
>> >> --
>> >> 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. Questions regarding this list
>> >> should be sent to stan AT temple DOT edu
>> >> =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
>> >
>> >--
>> >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. Questions regarding this list
>> >should be sent to stan AT temple DOT edu
>> >=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
>>
>> --
>> 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. Questions regarding this list
>> should be sent to stan AT temple DOT edu
>> =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
>
>--
>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. Questions regarding this list
>should be sent to stan AT temple DOT edu
>=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
--
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. Questions regarding this list
should be sent to stan AT temple DOT edu
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
|
