Re: [Gluster-infra] Netbsd folders filling up

[Emmanuel Dreyfus]

Nigel Babu  wrote:

> Thank you. I'll give that a shot. I also want to setup ntp

Add ntpd=YES in /etc/rc.conf and run /etc/rc.d/ntpd start
Note that the ntpdcurrently installed needs a secrity update.

> and change passwords for all the machines in one go.

Do it on one machine using vipw, copy /etc/master.passwd and run
pwd_mkdb -p /etc/master.passwd eveyrwhere to regenerate /etc/passwd 


-- 
Emmanuel Dreyfus
http://hcpnet.free.fr/pubz
m...@netbsd.org
___
Gluster-infra mailing list
Gluster-infra@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-infra

Re: [Gluster-infra] Netbsd folders filling up

[Nigel Babu]

Thank you. I'll give that a shot. I also want to setup ntp and change
passwords for all the machines in one go.

On Mon, Jul 18, 2016 at 5:22 PM, Emmanuel Dreyfus  wrote:

> Nigel Babu  wrote:
>
> > Oh, can I apply this to all the machines in one go?
>
> disklabel as is works with an interactive editor, but you can ealso
> disklabel xbd0 > protofile then tweak the file and use  disklabel -R
> xbd0 protofile to load is in a batch.
>
> Or you can just modify nbslave70, image it and deploy to other machines,
> it would not hurt.
>
> --
> Emmanuel Dreyfus
> http://hcpnet.free.fr/pubz
> m...@netbsd.org
>



-- 
nigelb
___
Gluster-infra mailing list
Gluster-infra@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-infra

Re: [Gluster-infra] Netbsd folders filling up

[Nigel Babu]

On Fri, Jul 15, 2016 at 03:57:39PM +0530, Nigel Babu wrote:
> On Fri, Jul 15, 2016 at 10:10:13AM +, Emmanuel Dreyfus wrote:
> > On Fri, Jul 15, 2016 at 03:04:39PM +0530, Nigel Babu wrote:
> > > Would it be okay to write a cron to clean up anything older than 15 days 
> > > in
> > > /build/install and /archives?
> >
> > You have to cleanup after some time. How is it handled on Linux boxen?
> >
> > --
> > Emmanuel Dreyfus
> > m...@netbsd.org
>
> They're all handled manually. There are fewer BSD boxes, so I'm guessing they
> fill up sooner. I'll put a cron for both linux and BSD similarly in one 
> ansible
> job. Or at least I hope to.
>
> --
> nigelb

Would it be problematic if I added 20GB of block storage per machine for the
/build, /home/jenkins and /archives folder? That should easily sort out our
disk space troubles.


--
nigelb
___
Gluster-infra mailing list
Gluster-infra@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-infra

Re: [Gluster-infra] Netbsd folders filling up

[Nigel Babu]

On Fri, Jul 15, 2016 at 10:10:13AM +, Emmanuel Dreyfus wrote:
> On Fri, Jul 15, 2016 at 03:04:39PM +0530, Nigel Babu wrote:
> > Would it be okay to write a cron to clean up anything older than 15 days in
> > /build/install and /archives?
>
> You have to cleanup after some time. How is it handled on Linux boxen?
>
> --
> Emmanuel Dreyfus
> m...@netbsd.org

They're all handled manually. There are fewer BSD boxes, so I'm guessing they
fill up sooner. I'll put a cron for both linux and BSD similarly in one ansible
job. Or at least I hope to.

--
nigelb
___
Gluster-infra mailing list
Gluster-infra@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-infra

Re: [Gluster-infra] Netbsd folders filling up

[Emmanuel Dreyfus]

On Fri, Jul 15, 2016 at 03:04:39PM +0530, Nigel Babu wrote:
> Would it be okay to write a cron to clean up anything older than 15 days in
> /build/install and /archives?

You have to cleanup after some time. How is it handled on Linux boxen?

-- 
Emmanuel Dreyfus
m...@netbsd.org
___
Gluster-infra mailing list
Gluster-infra@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-infra

Re: [Gluster-infra] Netbsd folders filling up

[Nigel Babu]

On Fri, Jul 15, 2016 at 07:22:45AM +, Emmanuel Dreyfus wrote:
> On Fri, Jul 15, 2016 at 07:19:17AM +, Emmanuel Dreyfus wrote:
> > On Fri, Jul 15, 2016 at 10:59:04AM +0530, Nigel Babu wrote:
> > > nbslave77.cloud.gluster.org
> >
> > That one has 1,6 Go of logs in /build/install/var/log/glusterfs
>
> And if you look for free space, you can wipe /usr/pkgsrc (2,6 Go)
> which is not used.
>
> --
> Emmanuel Dreyfus
> m...@netbsd.org

Would it be okay to write a cron to clean up anything older than 15 days in
/build/install and /archives?

--
nigelb
___
Gluster-infra mailing list
Gluster-infra@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-infra

Re: [Gluster-infra] Netbsd folders filling up

[Emmanuel Dreyfus]

On Fri, Jul 15, 2016 at 07:19:17AM +, Emmanuel Dreyfus wrote:
> On Fri, Jul 15, 2016 at 10:59:04AM +0530, Nigel Babu wrote:
> > nbslave77.cloud.gluster.org
> 
> That one has 1,6 Go of logs in /build/install/var/log/glusterfs

And if you look for free space, you can wipe /usr/pkgsrc (2,6 Go)
which is not used.

-- 
Emmanuel Dreyfus
m...@netbsd.org
___
Gluster-infra mailing list
Gluster-infra@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-infra

Re: [Gluster-infra] Netbsd folders filling up

[Nigel Babu]

So far the following are affected:

nbslave71.cloud.gluster.org
nbslave75.cloud.gluster.org
nbslave77.cloud.gluster.org
nbslave79.cloud.gluster.org
nbslave7g.cloud.gluster.org
nbslave7h.cloud.gluster.org
nbslave7j.cloud.gluster.org

I've cleaned out a few of them, but didn't get enough space. I'll take a
look at /var/crash in all of them.


On Fri, Jul 15, 2016 at 10:14 AM, Emmanuel Dreyfus  wrote:

> Nigel Babu  wrote:
>
> > Does anyone know which folders are usually filled up quickly on netbsd
> > machines? A lot of the machines are offline merely because they're out of
> > diskspace. I'm working through them. I've cleared out old files in /build
> > and /archives. Is there anywhere else I should be looking?
>
> What are the offending machines?
>
> Core files are configured to go in /var/crash, but /var/* should be a
> good place to look at.
>
> --
> Emmanuel Dreyfus
> http://hcpnet.free.fr/pubz
> m...@netbsd.org
>



-- 
nigelb
___
Gluster-infra mailing list
Gluster-infra@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-infra

Re: [Gluster-infra] Netbsd folders filling up

[Emmanuel Dreyfus]

Nigel Babu  wrote:

> Does anyone know which folders are usually filled up quickly on netbsd
> machines? A lot of the machines are offline merely because they're out of
> diskspace. I'm working through them. I've cleared out old files in /build
> and /archives. Is there anywhere else I should be looking?

What are the offending machines?

Core files are configured to go in /var/crash, but /var/* should be a
good place to look at.

-- 
Emmanuel Dreyfus
http://hcpnet.free.fr/pubz
m...@netbsd.org
___
Gluster-infra mailing list
Gluster-infra@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-infra