Pre Warewulf 3.7 question

[Jason Stover]

Hey all,

Does anyone know of any outstanding issues in the current SVN trunk,
or does it look good enough to everyone to kick 3.7 out the door?

-Jason

[Carlos Familia]

Hi,

In my opinion, 3.7.

Carlos

--
You received this message because you are subscribed to the Google Groups "Warewulf" group.
To unsubscribe from this group and stop receiving emails from it, send an email to warewulf+u...@lbl.gov.
To post to this group, send email to ware...@lbl.gov.
To view this discussion on the web visit https://groups.google.com/a/lbl.gov/d/msgid/warewulf/CAGfAqt8e_KtJbxt_zv4vKK5SHvPcAxbJ0MxGg-1q0O7MDZm3Uw%40mail.gmail.com.
For more options, visit https://groups.google.com/a/lbl.gov/d/optout.

[Allen, Benjamin S.]

The only thing from me would be the hwaddr based network interface configuration instead of device name based that we used today. This patch hasn't received much testing however. I can deal without it being in 3.7 depending on what everyone thinks.

Ben

[Jason Stover]

Hi Ben,

That patched _looked_ okay to me. So if you've been using it without
issue, you can commit it, or holler and I will.

-J

> --
> You received this message because you are subscribed to the Google Groups "Warewulf-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to warewulf-deve...@lbl.gov.

[Meij, Henk]

Does the disk repartitioning work?

-Henk

--

[Jason Stover]

Hi Henk,

I hate responding this way ... but I can't replicate it. So... "It
works for me"?

I'll see if I can't see about replicating it in a VM or something,
but none of the machines I've tested on have show the issue and will
change the partitioning...

-J

> https://groups.google.com/a/lbl.gov/d/msgid/warewulf/8688BD11DAC0574AA90295127E9E9F4AD77B028B%40EXCHANGEWES10.wesad.wesleyan.edu.

[Ryan Novosielski]

What was the issue? I'd be happy to try repartitioning on a couple of my clusters. We don't really use the disks for anything and have got CentOS 5/6/7 in use.

________________________________________
From: Jason Stover <jason....@gmail.com>
Sent: Friday, May 27, 2016 5:48 PM
To: ware...@lbl.gov
Cc: warewul...@lbl.gov
Subject: Re: [Warewulf] Pre Warewulf 3.7 question

To view this discussion on the web visit https://groups.google.com/a/lbl.gov/d/msgid/warewulf/CAGfAqt8F2gfVxwSk%3DjS6ZsnWj9c4Za22DAZE7-zYMR9DUcNHjg%40mail.gmail.com.

[Jason Stover]

Hi Ryan,

Basically, when provisioning, the disks were not being partitioned.
So, if a partition was already set on them, but a different
FILESYSTEMS value was set, and the node rebooted. The node still had
the old disk partition when it came up.

I... haven't been able to replicate it... and I can't see any reason
for it to happen with the 'dd' that should be wiping the partition
table out. :/

Thanks,
-J

> To view this discussion on the web visit https://groups.google.com/a/lbl.gov/d/msgid/warewulf/BY2PR14MB080614C74C348D410B3146CFFB420%40BY2PR14MB0806.namprd14.prod.outlook.com.

[Meij, Henk]

I can redo my tests next week

-Henk

To view this discussion on the web visit https://groups.google.com/a/lbl.gov/d/msgid/warewulf/CAGfAqt8F2gfVxwSk%3DjS6ZsnWj9c4Za22DAZE7-zYMR9DUcNHjg%40mail.gmail.com.

[Allen, Benjamin S.]

Committed.

I also found the postnetdown code around ifconfig is failing in VMs. When I run "ifconfig eth0 del 192.168.2.200 netmask 255.255.255.0" it returns "ifconfig: SIOCDIFADDR: Inappropriate ioctl for device".

If I have free time I'll debug this more.

Ben

[Allen, Benjamin S.]

Note, fixed below in r1989.

Ben

> You received this message because you are subscribed to the Google Groups "Warewulf" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to warewulf+u...@lbl.gov.

> To post to this group, send email to ware...@lbl.gov.

> To view this discussion on the web visit https://groups.google.com/a/lbl.gov/d/msgid/warewulf/0C61DFA7-2ACC-45DB-8F08-9F77B4047493%40alcf.anl.gov.

[Meij, Henk]

don't hold 3.7 because of this but I'm able to reproduce this behavior across nodes, I have info below.

 

when busybox comes up there is an error "Usage: switch_root -c ...." (this may be because mkbootable is SKIPPED after the getnfs process). I'd love to get this working.

 

-Henk

 

   Device Boot      Start         End      Blocks   Id  System
/dev/sda1   *           1          64      512000   83  Linux
Partition 1 does not end on cylinder boundary.
/dev/sda2              64        9730    77637632   8e  Linux LVM

 

[root@b1 ~]# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/mapper/vg_b1-lv_root
                       71G  3.8G   64G   6% /
/dev/sda1             485M   58M  402M  13% /boot

 

     158: NAME       = b1
     158: BOOTLOADER = sda
     158: BOOTSTRAPID = 6
     158: DISKFORMAT = sda1,sda3
     158: FILEIDS    = 16,19,2,20,22,4,5
     158: FILESYSTEMS = dev=sda2:type=swap:size=12048,mountpoint=/:dev=sda3:type=ext4:size=fill,mountpoint=/boot:dev=sda1:type=ext4:size=500

---

From: Meij, Henk [hm...@wesleyan.edu]
Sent: Friday, May 27, 2016 6:15 PM
To: ware...@lbl.gov
Cc: warewul...@lbl.gov
Subject: Re: [Warewulf] Pre Warewulf 3.7 question

To view this discussion on the web visit https://groups.google.com/a/lbl.gov/d/msgid/warewulf/8688BD11DAC0574AA90295127E9E9F4AD77B0348%40EXCHANGEWES10.wesad.wesleyan.edu.

[Jason Stover]

Well, I'd like to kill this issue if we can find what's causing it.
What drive model is it you're having an issue with?

-J

> https://groups.google.com/a/lbl.gov/d/msgid/warewulf/8688BD11DAC0574AA90295127E9E9F4AD77B64F1%40EXCHANGEWES10.wesad.wesleyan.edu.

[Meij, Henk]

I can produce this error with

HP 160G SATA model GB0160EAPRR
Western Digital WDC WD800BB-75H

-Henk

________________________________________
From: Jason Stover [jason....@gmail.com]

Sent: Thursday, June 02, 2016 4:19 PM

To view this discussion on the web visit https://groups.google.com/a/lbl.gov/d/msgid/warewulf/CAGfAqt_wR_y53eL0FtVHeYhBmXcncs1JZ2KQwVT%3DMY5OLYJYYw%40mail.gmail.com.

[Jason Stover]

Well, for the HP drive, the only possible issue I can find that
_could_ possibly be causing it, would be a firmware issue where the
drive would send stale data... if that includes the partition table
.... I don't know. :/

Here's the doc:
ftp://ftp.hp.com/pub/softlib2/software1/doc/p1822529277/v76998/SPP2012.02.0BReleaseNotes.pdf

Of course, the linked to page in the section results in nothing being found....

I'm still going to see if I can't find some way to replicate this in a
VM. If I can't, then I'll go ahead and branch off 3.7 and get gmk to
make the tarballs.

-J

> To view this discussion on the web visit https://groups.google.com/a/lbl.gov/d/msgid/warewulf/8688BD11DAC0574AA90295127E9E9F4AD77B864E%40EXCHANGEWES10.wesad.wesleyan.edu.

[Meij, Henk]

ah, I see what you're thinking, perhaps not a warewulf issue at all, yes this is all very old hardware and not maintained (>7 years) and repurposed, and useable using existing partitions

if I get my hands on new hardware I'll test it out ... with 3.7 ;)

-Henk

________________________________________
From: Jason Stover [jason....@gmail.com]

Sent: Friday, June 03, 2016 11:01 AM

To view this discussion on the web visit https://groups.google.com/a/lbl.gov/d/msgid/warewulf/CAGfAqt8nM7T88e1-40x5jXZqspPfXK4sZtdSPkRw4EmEAh9qxg%40mail.gmail.com.

[Jason Stover]

Hi Henk,

Finally got my stupid VM's up... and I still don't have an issue
with the partitioning. Have modified a node so that it's disk
partitioning has changed multiple times, etc... and the partition
table gets written out correctly.

So, on the note, I'm going to go a head and get 3.7 branched so GMK
can push packages out the door. If you do run into something that you
trace back to how Warewulf is doing things, then holler at us.

-J