warning Listener::getElapsedTimeMillis returning 0ms

[Roy Smith]

I'm doing a mongorestore of a database dump.  I'm running Mongo 2.0.2 on Ubuntu Linux.  I'm getting a large number of messages:

Wed Jan 25 18:08:33 [tools] warning Listener::getElapsedTimeMillis returning 0ms

Wed Jan 25 18:08:34 [tools] warning Listener::getElapsedTimeMillis returning 0ms

Any idea why this is happening or what it means, operationally?

PS -- the restore finished, and seems to be OK.  But the long stream of warnings is unsettling.

This is the first time I've done a restore with 2.0.  Previously, we were using 1.6 or 1.8.  Interestingly enough, restores used to take about 12 hours (50-ish GB of data).  This completed in about 4 hours!  Yeah!

[Scott Hernandez]

How did you install the package and where is log from, the server or client?

> --
> You received this message because you are subscribed to the Google Groups
> "mongodb-user" group.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/mongodb-user/-/WDGclk_KOFMJ.
> To post to this group, send email to mongod...@googlegroups.com.
> To unsubscribe from this group, send email to
> mongodb-user...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/mongodb-user?hl=en.

[Roy Smith]

Neither server nor client.  I was running mongorestore in stand-alone mode...

$ mongorestore --dbpath 2012-01-25 /mnt/roy/songza/db_backups/mongo-all-2012-01-25

It was installed via a debian package.  You can grab the package from https://bitbucket.org/wesc/debian-mongodb/

[Scott Hernandez]

How was this compiled/build? So this is not from the official repo, or
builds, here? http://www.mongodb.org/display/DOCS/Ubuntu+and+Debian+packages

If you do this with an online server, not using dbpath, do you get
this error in the server or tool output/logs?

> --
> You received this message because you are subscribed to the Google Groups
> "mongodb-user" group.
> To view this discussion on the web visit

> https://groups.google.com/d/msg/mongodb-user/-/tDkGNnq5Uq4J.

[Roy Smith]

I have not tried doing a restore via the server.  We always do restores directly to disk.

[Wes Chow]

The package is based on Debian's 2.0.2-1 package (not 10gen's), but is slightly modified to build on Ubuntu Lucid. The packaging does not do much except fix sconstruct problems and install / link paths. You can see the patches here:

https://bitbucket.org/wesc/debian-mongodb/src/961bf3ffb8c4/mongodb-2.0.2/debian/patches

I'm unaware of any changes Debian may have made to the package beyond what is in that patch directory.

Wes

[Scott Hernandez]

From my experience there is little to no advantage to doing that than
using a running instance. If you plan on running the server and only
stop it to do this, I would suggest just running this as an online
operation.

On Thu, Jan 26, 2012 at 12:13 PM, Roy Smith <r...@panix.com> wrote:
> I have not tried doing a restore via the server.  We always do restores
> directly to disk.
>

> --
> You received this message because you are subscribed to the Google Groups
> "mongodb-user" group.
> To view this discussion on the web visit

> https://groups.google.com/d/msg/mongodb-user/-/lV1VJ0xm7twJ.

[Wes Chow]

We started restoring directly to disk when we were on a 1.6.x version, and at the time it was significantly faster. For the size of our database, often faster by several hours. We never looked back, so perhaps you're correct that it's no longer advantageous from a speed perspective.

Another nice thing about restoring directly to disk is that you can minimize mongod downtime. Our use case is this:

1. mongodump on production daily and archive.

2. Every so often, a developer will download an archived dump and restore on his local desktop.

It would be nice if #2 can happen while mongod is still up and serving consistent data, so the developer can still do work.

That said, in practice these restores are kicked off as the last thing the developer does for the day, and they're complete by the morning. Pre 2.0 they took about 6 hours, and would essentially make the machine unusable.

Wes

[Ian Wood]

So.. Does anybody know what these warnings are about and why?

[Asya Kamsky]

These were pretty long time ago - I would hope no one is still using
2.0.2 given production is up to 2.4.5 now...

Asya

> --

> --
> You received this message because you are subscribed to the Google
> Groups "mongodb-user" group.

> To post to this group, send email to mongod...@googlegroups.com

> To unsubscribe from this group, send email to
> mongodb-user...@googlegroups.com

> See also the IRC channel -- freenode.net#mongodb
>
> ---

> You received this message because you are subscribed to the Google Groups
> "mongodb-user" group.

> To unsubscribe from this group and stop receiving emails from it, send an
> email to mongodb-user...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

[Tom Boutell]

This is an old thread, but I just got these while running:

mongod --dbpath /usr/local/var/mongodb --repair

I have 2.6.2.

The message appears at regular intervals during the repair, which eventually succeeds.

[Asya Kamsky]

It's nothing to worry about - has to do with commands tracking elapsed
time and I guess repair() doesn't so there is a log line (which should
only be printed rarely so as not to fill the log with it) that
mentions that.

Asya

> For other MongoDB technical support options, see:
> http://www.mongodb.org/about/support/.

> ---
> You received this message because you are subscribed to the Google Groups
> "mongodb-user" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to mongodb-user...@googlegroups.com.

> To post to this group, send email to mongod...@googlegroups.com.
> Visit this group at http://groups.google.com/group/mongodb-user.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/mongodb-user/30c84262-872b-4a3e-8ae1-7408556e7620%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.