PXE boot halt after 1.5.2 upgrade

[Hongyun]

I upgrade our FOG server to the 1.5.2 version today, but after the upgrade, PXE boot halt, and won’t pass to start the computer.

Attached image is what I see on the client screen.

Here is the message log on the server:
May 14 14:02:21 foggy kernel: FWALL: IN=eth0 OUT= MAC=ff:ff:ff:ff:ff:ff:70:71:bc:1f:ff:b6:08:00 SRC=0.0.0.0 DST=255.255.255.255 LEN=576 TOS=0x00 PREC=0x00 TTL=20 ID=1 PROTO=UDP SPT=68 DPT=67 LEN=556
May 14 14:02:23 foggy kernel: FWALL: IN=eth0 OUT= MAC=ff:ff:ff:ff:ff:ff:70:71:bc:1f:ff:b6:08:00 SRC=0.0.0.0 DST=255.255.255.255 LEN=576 TOS=0x00 PREC=0x00 TTL=20 ID=2 PROTO=UDP SPT=68 DPT=67 LEN=556
May 14 14:02:23 foggy xinetd[1919]: START: tftp pid=6474 from=142.103.243.44
May 14 14:02:23 foggy in.tftpd[6475]: tftp: client does not accept options
May 14 14:02:23 foggy in.tftpd[6476]: Client 142.103.243.44 finished undionly.kpxe
May 14 14:02:28 foggy kernel: FWALL: IN=eth0 OUT= MAC=ff:ff:ff:ff:ff:ff:70:71:bc:1f:ff:b6:08:00 SRC=0.0.0.0 DST=255.255.255.255 LEN=437 TOS=0x00 PREC=0x00 TTL=64 ID=256 PROTO=UDP SPT=68 DPT=67 LEN=417
May 14 14:02:29 foggy kernel: FWALL: IN=eth0 OUT= MAC=ff:ff:ff:ff:ff:ff:70:71:bc:1f:ff:b6:08:00 SRC=0.0.0.0 DST=255.255.255.255 LEN=437 TOS=0x00 PREC=0x00 TTL=64 ID=549 PROTO=UDP SPT=68 DPT=67 LEN=417
May 14 14:02:35 foggy kernel: FWALL: IN=eth0 OUT= MAC=ff:ff:ff:ff:ff:ff:70:71:bc:1f:ff:b6:08:00 SRC=0.0.0.0 DST=255.255.255.255 LEN=449 TOS=0x00 PREC=0x00 TTL=64 ID=2018 PROTO=UDP SPT=68 DPT=67 LEN=429
May 14 14:02:35 foggy in.tftpd[6485]: Client 142.103.243.44 finished default.ipxe

Does anyone know what might be the problem? This is affecting all our lab computers now, if I reboot any of them it will stop on this screen .

Thanks,
Hongyun

[Wayne Workman]

Does it only affect the dual booting systems?

[Hongyun]

No, all the systems. We have computers with only Linux or dual boot.

[Wayne Workman]

@hongyun Check the apache error log, see if you find any errors from the time when you’re trying to network boot a system.

[Hongyun]

@wayne-workman Here are the httpd error log:

[Mon May 14 14:00:51 2018] [notice] suEXEC mechanism enabled (wrapper: /usr/sbin/suexec)
[Mon May 14 14:00:51 2018] [notice] Digest: generating secret for digest authentication …
[Mon May 14 14:00:51 2018] [notice] Digest: done
[Mon May 14 14:00:51 2018] [notice] Apache/2.2.15 (Unix) DAV/2 mod_auth_tkt/2.1.3 (Opsview) mod_ssl/2.2.15 OpenSSL/1.0.1e-fips mod_wsgi/3.2 Python/2.6.6 mod_perl/2.0.4 Perl/v5.10.1 configured – resuming normal operations
[Mon May 14 14:00:51 2018] [error] avahi_entry_group_add_service_strlst(“142.103.243.243”) failed: Invalid host name
BFD: /var/www/html/fog//service/ipxe/bzImage32: Warning: Ignoring section flag IMAGE_SCN_MEM_NOT_PAGED in section .bss

[Wayne Workman]

@developers @moderators anyone see this before?

[george1421]

@hongyun It appears that from your initial post that the linux firewall is turned on some how.

May 14 14:02:21 foggy kernel: FWALL: IN=eth0 OUT= MAC=ff:ff:ff:ff:ff:ff:70:71:bc:1f:ff:b6:08:00 SRC=0.0.0.0 DST=255.255.255.255 LEN=576 TOS=0x00 PREC=0x00 TTL=20 ID=1 PROTO=UDP SPT=68 DPT=67 LEN=556

This message is basically saying that a dhcp request is being blocked by the firewall. One of the prerequisites for installing fog is to disable the linux firewall on the fog server as well as setting selinux to permissive. Without these two settings in place fog will not work.

[george1421]

@hongyun Is this FOG server doing more functions than we think?

I see OpsView and avahi messages in the log.

[Hongyun]

We have a separate DHCP server, the fog server only provide PXE boot. I can try disable firewall. But because I don’t want to affect the lab usage, I have already roll back to the 1.4.0 version, which works fine with our current settings.

[Wayne Workman]

I’d recommend running nothing except FOG on your FOG server. We’re not able to troubleshoot integration with other 3rd party software, that’s on you.