All of lore.kernel.org
 help / color / mirror / Atom feed
* Dracut NFS root DHCP timeouts
@ 2010-10-11  2:37 daniel-2N1FInP7wozQT0dZR+AlfA
       [not found] ` <20101011133718.205549hmcnjzr3a8-2RFepEojUI3Y3UUic/vCjNBPR1lH4CV8@public.gmane.org>
  0 siblings, 1 reply; 4+ messages in thread
From: daniel-2N1FInP7wozQT0dZR+AlfA @ 2010-10-11  2:37 UTC (permalink / raw)
  To: nitramfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org

Hi Guys,

I'm getting their with my Dracut setup however I'm having a weird DHCP issue.

I can't get DHCP 4.0.1 or 4.1.0 working with Gentoo and Dracut 007.

I've downgraded to DHCP 3.1.2p1 and removed the -4 flag from the  
Dracut scripts.

The PXEBoot line is:
APPEND bootdev=eth1 ip:eth0:off ip=eth1:dhcp  
root=10.197.170.5:/vol/filerboot/linux/images/live-mon-test:fg,timeo=600,retrans=2,rsize=32768,wsize=32768,hard,nfsvers=3,nointr,actimeo=3600,noacl,rw,tcp,sec=sys initrd=initrd-2.6.34-gentoo-r1-dracut  
rw

While the server boots and the interfaces initializers I see:
Sending DHCP requets .....timed out

The following happens 2-3 times

The DHCP server logs report..

Attempt 1:
dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:4e  
via 10.197.220.241: ignored (not authoritative).
dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:50  
via eth1: ignored (not authoritative).
dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:4e  
via 10.197.220.241: ignored (not authoritative).
dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:50  
via eth1: ignored (not authoritative).
dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:4e  
via 10.197.220.241: ignored (not authoritative).
dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:50  
via eth1: ignored (not authoritative).
dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:4e  
via 10.197.220.241: ignored (not authoritative).
dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:50  
via eth1: ignored (not authoritative).

Attempts 2:
dhcpd: DHCPREQUEST for 255.255.255.255 (180.0.0.0) from  
00:19:b9:eb:3d:4e via 10.197.220.241: ignored (not authoritative).
dhcpd: DHCPREQUEST for 255.255.255.255 (180.0.0.0) from  
00:19:b9:eb:3d:50 via eth1: ignored (not authoritative).
dhcpd: DHCPREQUEST for 255.255.255.255 (180.0.0.0) from  
00:19:b9:eb:3d:4e via 10.197.220.241: ignored (not authoritative).
dhcpd: DHCPREQUEST for 255.255.255.255 (180.0.0.0) from  
00:19:b9:eb:3d:50 via eth1: ignored (not authoritative).
dhcpd: DHCPREQUEST for 255.255.255.255 (180.0.0.0) from  
00:19:b9:eb:3d:4e via 10.197.220.241: ignored (not authoritative).
dhcpd: DHCPREQUEST for 255.255.255.255 (180.0.0.0) from  
00:19:b9:eb:3d:50 via eth1: ignored (not authoritative).

Just before the pivot DHCP runs again and all works. The DHCP server reports:

dhcpd: DHCPDISCOVER from 00:19:b9:eb:3d:50 via eth1
dhcpd: DHCPOFFER on 10.197.170.30 to 00:19:b9:eb:3d:50 via eth1
dhcpd: Dynamic and static leases present for 10.197.170.30.
dhcpd: Remove host declaration san.n15-pizza1-3 or remove 10.197.170.30
dhcpd: from the dynamic address pool for 10.197.170.0/24
dhcpd: DHCPREQUEST for 10.197.170.30 (10.197.170.42) from  
00:19:b9:eb:3d:50 via eth1
dhcpd: DHCPACK on 10.197.170.30 to 00:19:b9:eb:3d:50 via eth1

Gentoo boots and all seems to be ok except for a weird issue where the  
resolv.conf is not updated correctly.


Any ideas on what the causing the initial DHCP timeouts? It makes the  
boot process take a lot longer!


Thanks




^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Dracut NFS root DHCP timeouts
       [not found] ` <20101011133718.205549hmcnjzr3a8-2RFepEojUI3Y3UUic/vCjNBPR1lH4CV8@public.gmane.org>
@ 2010-10-11 12:25   ` Seewer Philippe
       [not found]     ` <4CB30231.80803-omB+W0Dpw2o@public.gmane.org>
  0 siblings, 1 reply; 4+ messages in thread
From: Seewer Philippe @ 2010-10-11 12:25 UTC (permalink / raw)
  To: daniel-2N1FInP7wozQT0dZR+AlfA@public.gmane.org
  Cc: <initramfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>

Hi Daniel

On 10/11/2010 04:37 AM, daniel-2N1FInP7wozQT0dZR+AlfA@public.gmane.org wrote:
> While the server boots and the interfaces initializers I see:
> Sending DHCP requets .....timed out
>
> The following happens 2-3 times
>
> The DHCP server logs report..
>
> Attempt 1:
> dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:4e
> via 10.197.220.241: ignored (not authoritative).
> dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:50
> via eth1: ignored (not authoritative).
> dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:4e
> via 10.197.220.241: ignored (not authoritative).
> dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:50
> via eth1: ignored (not authoritative).
> dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:4e
> via 10.197.220.241: ignored (not authoritative).
> dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:50
> via eth1: ignored (not authoritative).
> dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:4e
> via 10.197.220.241: ignored (not authoritative).
> dhcpd: DHCPREQUEST for 201.0.0.0 (180.0.0.0) from 00:19:b9:eb:3d:50
> via eth1: ignored (not authoritative).
>
> Attempts 2:
> dhcpd: DHCPREQUEST for 255.255.255.255 (180.0.0.0) from
> 00:19:b9:eb:3d:4e via 10.197.220.241: ignored (not authoritative).
> dhcpd: DHCPREQUEST for 255.255.255.255 (180.0.0.0) from
> 00:19:b9:eb:3d:50 via eth1: ignored (not authoritative).
> dhcpd: DHCPREQUEST for 255.255.255.255 (180.0.0.0) from
> 00:19:b9:eb:3d:4e via 10.197.220.241: ignored (not authoritative).
> dhcpd: DHCPREQUEST for 255.255.255.255 (180.0.0.0) from
> 00:19:b9:eb:3d:50 via eth1: ignored (not authoritative).
> dhcpd: DHCPREQUEST for 255.255.255.255 (180.0.0.0) from
> 00:19:b9:eb:3d:4e via 10.197.220.241: ignored (not authoritative).
> dhcpd: DHCPREQUEST for 255.255.255.255 (180.0.0.0) from
> 00:19:b9:eb:3d:50 via eth1: ignored (not authoritative).
>
> Just before the pivot DHCP runs again and all works. The DHCP server reports:
>
> dhcpd: DHCPDISCOVER from 00:19:b9:eb:3d:50 via eth1
> dhcpd: DHCPOFFER on 10.197.170.30 to 00:19:b9:eb:3d:50 via eth1
> dhcpd: Dynamic and static leases present for 10.197.170.30.
> dhcpd: Remove host declaration san.n15-pizza1-3 or remove 10.197.170.30
> dhcpd: from the dynamic address pool for 10.197.170.0/24
> dhcpd: DHCPREQUEST for 10.197.170.30 (10.197.170.42) from
> 00:19:b9:eb:3d:50 via eth1
> dhcpd: DHCPACK on 10.197.170.30 to 00:19:b9:eb:3d:50 via eth1
>
> Gentoo boots and all seems to be ok except for a weird issue where the
> resolv.conf is not updated correctly.

Ummm... Just so that I understand this. The logs for Attempt 1 and 2 are 
after the dracut pivot from the dhcp requests inside the nfs-rooted 
system and the logs and the last one above is from dhcp inside dracut. 
correct?

> Any ideas on what the causing the initial DHCP timeouts? It makes the
> boot process take a lot longer!

BTW: If you are using a recent pxelinux, why don't you try IPAPPEND=3 
and remove all ip-options from dracut? (See 
http://syslinux.zytor.com/wiki/index.php/SYSLINUX#IPAPPEND_flag_val_.5BPXELINUX_only.5D). 
This should cause dracut not to issue any dhcp requests at all. Sorry I 
haven't thought about this earlier.

Regards,
Philippe

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Dracut NFS root DHCP timeouts
       [not found]     ` <4CB30231.80803-omB+W0Dpw2o@public.gmane.org>
@ 2010-10-11 22:44       ` daniel-2N1FInP7wozQT0dZR+AlfA
       [not found]         ` <20101012094412.14125sp7tc6dj7r4-2RFepEojUI3Y3UUic/vCjNBPR1lH4CV8@public.gmane.org>
  0 siblings, 1 reply; 4+ messages in thread
From: daniel-2N1FInP7wozQT0dZR+AlfA @ 2010-10-11 22:44 UTC (permalink / raw)
  To: Seewer Philippe; +Cc: <initramfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>

Thanks Seewer!

Quoting Seewer Philippe <philippe.seewer-omB+W0Dpw2o@public.gmane.org>:

> Ummm... Just so that I understand this. The logs for Attempt 1 and 2  
> are after the dracut pivot from the dhcp requests inside the  
> nfs-rooted system and the logs and the last one above is from dhcp  
> inside dracut. correct?

All attempts are pre-pivot. The first 2 happen early in boot and then  
the final appears to happen as the final step  prior to the pivot to  
the NFS root.

> BTW: If you are using a recent pxelinux, why don't you try  
> IPAPPEND=3 and remove all ip-options from dracut? (See  
> http://syslinux.zytor.com/wiki/index.php/SYSLINUX#IPAPPEND_flag_val_.5BPXELINUX_only.5D). This should cause dracut not to issue any dhcp requests at all. Sorry I haven't thought about this  
> earlier.

I tried with IPAPPEND=3 and Dracut fails to boot. It appears to get an  
IP but dies with an error:

[snip]
IP-Config: Got DHCP answer from 10.197.170.42, my address is 10.197.170.30
IP-Config: Complete:
        device=eth1, addr=10.197.170.30, mask=255.255.255.0,  
gw=255.255.255.255,
        host=test.test.com, domain=test.com, nis-domain=(none),
        bootserver=10.197.170.42, rootserver=10.197.170.42, rootpath=
Freeing unused kernel memory: 452k freed
dracut: FATAL: Mixing BOOTIF and ip= lines is dangerous
dracut: Refusing to continue
dracut: FATAL: Mixing BOOTIF and ip= lines is dangerous
dracut: Refusing to continue


Signal caught!

Boot has failed, sleeping forever
[/snip]

The initial DHCP request no longer occur, dracut appears to get an  
address however it then fails to pivot?




^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Dracut NFS root DHCP timeouts
       [not found]         ` <20101012094412.14125sp7tc6dj7r4-2RFepEojUI3Y3UUic/vCjNBPR1lH4CV8@public.gmane.org>
@ 2010-10-11 22:59           ` daniel-2N1FInP7wozQT0dZR+AlfA
  0 siblings, 0 replies; 4+ messages in thread
From: daniel-2N1FInP7wozQT0dZR+AlfA @ 2010-10-11 22:59 UTC (permalink / raw)
  To: Seewer Philippe,
	<initramfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>


> Quoting Seewer Philippe <philippe.seewer-omB+W0Dpw2o@public.gmane.org>:

>> BTW: If you are using a recent pxelinux, why don't you try  
>> IPAPPEND=3 and remove all ip-options from dracut? (See

I tried IPAPPEND=2 and everything works correctly. The initial DHCP  
requests do not occur and then prior to pivot it gets the address on  
the boot interface and boots!

So the pxelinux.cfg looks like:

                 IPAPPEND 2 # initially tried 3 without success
                 APPEND  
root=10.197.170.5:/vol/filerboot/linux/images/live-mon-test:fg,timeo=600,retrans=2,rsize=32768,wsize=32768,hard,nfsvers=3,nointr,actimeo=3600,noacl,rw,tcp,sec=sys initrd=initrd-2.6.34-gentoo-r1-dracut  
rw



^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2010-10-11 22:59 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-10-11  2:37 Dracut NFS root DHCP timeouts daniel-2N1FInP7wozQT0dZR+AlfA
     [not found] ` <20101011133718.205549hmcnjzr3a8-2RFepEojUI3Y3UUic/vCjNBPR1lH4CV8@public.gmane.org>
2010-10-11 12:25   ` Seewer Philippe
     [not found]     ` <4CB30231.80803-omB+W0Dpw2o@public.gmane.org>
2010-10-11 22:44       ` daniel-2N1FInP7wozQT0dZR+AlfA
     [not found]         ` <20101012094412.14125sp7tc6dj7r4-2RFepEojUI3Y3UUic/vCjNBPR1lH4CV8@public.gmane.org>
2010-10-11 22:59           ` daniel-2N1FInP7wozQT0dZR+AlfA

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.