Console Upgrade from 1.2 Release Embeded -> 1.2.1 BETA Embeded [Now Working]

cheesyboofs

Expanded a new 512Mb image and injected a config file then did a fresh write to my dev micro drive and all is cooking - so far.

Will try an upgrade to a newer snapshot to see if that works tomorrow. Would be a right pain in the arse if I had to build a new image each time.

Keep up the fantastic work and thanks for making the snapshots public.

Cheesy.

celtic

tried upgrading too, it hung on booting just beore loading the kernel…
installed a resh install... ( no embedded... ) that booted at least...

sullrich

Try using the Full update file from the console? Does it work better on upgrades?

cheesyboofs

YOU STAR!

While we wait for the DHCP fix I dropped back to 1.2 RELEASE then upgraded from console with -

http://snapshots.pfsense.org/FreeBSD7/RELENG_1_2/_updaters/latest.tgz

(07-Jul-2008 21:10 - 38M)

And what do you know, it worked!

When I logged in to the GUI for the first time I saw something I have never seen before some package installer thing that then said all packages re-installed? I don't have packages installed as we all know "their not supported" :P

Oh and DHCP is now working CONFIRMED!


      ___
 ___/ f \
/ p \___/ Sense
\___/   \
    \___/

Welcome to pfSense 1.2.1-TESTING-SNAPSHOT on the 'embedded' platform...

Setting up embedded specific environment... done.
Mounting filesystems... done.
Creating symlinks......done.
Launching PHP init system... done.
Initializing.................. done.
Starting device manager (devd)...done.
Loading configuration......done.
Updating configuration...done.
Cleaning backup cache...done.
Setting up extended sysctls...done.
Syncing user passwords...done.
Starting Secure Shell Services...done.
Setting timezone...done.
Starting syslog...done.
Configuring LAN interface...done.
Configuring WAN interface...

cheesyboofs

I think LoadBalance is broke!

My working config now looks like this - Don't know how I'm even on the internet :o

loadbalance3.jpg_thumb

loadbalance.jpg_thumb
loadbalance2.jpg_thumb

sullrich

Is slbd running? ps awux | grep slbd from a shell.

If not, run slbd and return its output.

sullrich

Found the problem with fping. Please try the next snapshot that will be generated a few hours from now.

cheesyboofs

Right tried remote upgrade via ssh with file,

http://snapshots.pfsense.org/FreeBSD7/RELENG_1_2/_updaters/latest.tgz

which at the time was 1.2.1 (08-Jul-2008 04:02 38M) can confirm LoadBalance is now working but can also confirm that DHCP is broken again.

untitled.jpg_thumb

sullrich

DHCP is now fixed. Check next snap.

cheesyboofs

Hi sullrich,

Right first I dropped back to 1.2-Release on my dev micro vault then upgraded to 1.2.1 BETA with the embedded update with the following result,

pfSense-Embedded-Update-1.2.1-TESTING-SNAPSHOT-20080709-0101.tgz

Enter the URL to the .tgz update file:
> http://snapshots.pfsense.org/FreeBSD7/RELENG_1_2/pfSense-Embedded-Update-1.2.1-TESTING-SNAPSHOT-20080709-0101.tgz

Fetching file size...

File size: 23283897

Fetching file...
looking up snapshots.pfsense.org
connecting to snapshots.pfsense.org:80
requesting http://snapshots.pfsense.org/FreeBSD7/RELENG_1_2/pfSense-Embedded-Update-1.2.1-TESTING-SNAPSHOT-20080709-0101.tgz
remote size / mtime: 23283897 / 1215567318
/root/firmware.tgz                            100% of   22 MB  122 kBps 00m00s

Fetching MD5...
looking up snapshots.pfsense.org
connecting to snapshots.pfsense.org:80
requesting http://snapshots.pfsense.org/FreeBSD7/RELENG_1_2/pfSense-Embedded-Update-1.2.1-TESTING-SNAPSHOT-20080709-0101.tgz.md5
remote size / mtime: 128 / 1215568077
/root/firmware.tgz.md5                        100% of  128  B  287  Bps
URL MD5: 6d43cdbee47bc6f5daaf90b565499ba8

Downloaded file MD5: 6d43cdbee47bc6f5daaf90b565499ba8

MD5 checksum matches.

One moment please... Invoking firmware upgrade...

*** FINAL System shutdown message from root@router.cheesyboofs.co.uk ***
System going down IMMEDIATELY


/boot.config: -D
Consoles: internal video/keyboard  serial port
BIOS drive C: is disk0
BIOS 638kB/980864kB available memory

FreeBSD/i386 bootstrap loader, Revision 1.1
(sullrich@freebsd7-releng_1_2.geekgod.com, Wed Jul  9 00:17:30 EDT 2008)
Loading /boot/defaults/loader.conf
Unable to load a kernel!
-
can't load 'kernel'

Type '?' for a list of commands, 'help' for more detailed help.
OK

So I then went back to using the full update,

pfSense-Full-Update-1.2.1-TESTING-SNAPSHOT-20080709-0101.tgz

Enter the URL to the .tgz update file:
> http://snapshots.pfsense.org/FreeBSD7/RELENG_1_2/pfSense-Full-Update-1.2.1-TESTING-SNAPSHOT-20080709-0101.tgz

Fetching file size...

File size: 39623943

Fetching file...
looking up snapshots.pfsense.org
connecting to snapshots.pfsense.org:80
requesting http://snapshots.pfsense.org/FreeBSD7/RELENG_1_2/pfSense-Full-Update-1.2.1-TESTING-SNAPSHOT-20080709-0101.tgz
remote size / mtime: 39623943 / 1215567593
/root/firmware.tgz                            100% of   37 MB  159 kBps 00m00s

Fetching MD5...
looking up snapshots.pfsense.org
connecting to snapshots.pfsense.org:80
requesting http://snapshots.pfsense.org/FreeBSD7/RELENG_1_2/pfSense-Full-Update-1.2.1-TESTING-SNAPSHOT-20080709-0101.tgz.md5
remote size / mtime: 124 / 1215568077
/root/firmware.tgz.md5                        100% of  124  B  311  Bps
URL MD5: 7cf5f9f397051d18e847bd6cfc07afac

Downloaded file MD5: 7cf5f9f397051d18e847bd6cfc07afac

MD5 checksum matches.

One moment please... Invoking firmware upgrade...

*** FINAL System shutdown message from root@router.cheesyboofs.co.uk ***
System going down IMMEDIATELY

Now everything I use my pfSense box for seems to be working great, DHCP, Loadbalance/Fail-over, OpenVPN and traffic-shaping, so many thanks for all your hard work.

You might as well stop compiling the embedded update files for a while until you get time to fix them because they just don't work.

Please don't give up on embedded altogether as I would be gutted if I had to stop using my Sony micro vaults.
They are so handy to have a disaster recovery spare standing by and they only take 60 seconds to write a new one - I love um'

cheesyboofs

Well she ran solid for four hours then locked up completely! :-\

I have had to drop back to 1.2-Release to remain stable until I can do some more testing. If its gunna take 4 hours each time it could take a while.
The hardware is know good and runs fine on 1.2 so this must be a problem with drivers in v7.0 or the fact that I had to use the 'full' update on an embedded install.
As I am running embedded no log files exist as to why the lockup occurred.

cheesyboofs

Unfortunately last night I had another lockup but this time it lasted a full week before the lockup occurred, so I have dropped back once again to 1.2-RELEASE until at least someone else reports a lockup and I can compare notes.

eri--

Can you even give details about your hardware i am not seeing anywhere!

Ermal

cheesyboofs

A Jetway J7F4K 1.2Ghz + 1Gb Ram

http://www.jetway.com.tw/jw/ipcboard_view.asp?productid=279&proname=J7F4K1G2E

a D-Link DFE-580TX Quad NIC PCI Card,

http://linitx.com/viewproduct.php?prodid=10393

Together run like an absolute dream under 1.2-Release embedded but has locked up twice under two different builds of 1.2.1 BETA.

I will run for a month on 1.2 to make sure a fault hasn't crept in but I'm sure its all fine as there is little that can go wrong really.

Cheers

index2.jpg_thumb

router.jpg_thumb

cheesyboofs

Well 15 Day's back on 1.2-Release without error - I clearly don't have a hardware fault.

I will now try 1.2.1 again and pray it doesn't lock-up again.

Untitled-1.jpg_thumb

cheesyboofs

Yet another lockup after being up for 5 days - this clearly is a compatibility problem between FreeBSD 7 and my hardware or a result of upgrading from embedded 1.2 to 1.2.1-BETA with the FULL update.

To try and rule out FreeBSD 7, I am now reluctantly going back to a non-embedded full install of 1.2.1 to a CF II Microdrive.

eri--

Can you be more specific on the meaning of the 'lockup' word?

cheesyboofs

I mean lockup in the old fashioned sence of the word:

No network activity on any interface (I have four active, 2 WAN, 1 Wireless DMZ and 1 LAN)
No serial console access.
No controlled halt when the power button is depressed.

I set up remote sys logging in the hope I may catch something as when it happens I loose any history for obvious reasons but alas it didn't show anything other than the time it locked up and the following reboot when I got home from work:

2008-08-12 16:43:44 Local0.Info 192.168.100.254 Aug 12 16:43:44 pf: 86. 848637 rule 142/0(match): pass in on ste1: (tos 0x0, ttl 110, id 27768, offset 0, flags [DF], proto TCP (6), length 48) 212.92.4.173.64893 > 192.168.100.5.25: [|tcp]
2008-08-12 16:44:34 Local0.Info 192.168.100.254 Aug 12 16:44:34 pf: 49. 779067 rule 142/0(match): pass in on ste1: (tos 0x0, ttl 114, id 37811, offset 0, flags [DF], proto TCP (6), length 48) 59.182.35.69.7099 > 192.168.100.5.25: [|tcp]
2008-08-12 16:45:17 Local0.Info 192.168.100.254 Aug 12 16:45:17 pf: 43. 584296 rule 142/0(match): pass in on ste1: (tos 0x0, ttl 114, id 6379, offset 0, flags [DF], proto TCP (6), length 48) 59.182.35.69.7244 > 192.168.100.5.25: [|tcp]
2008-08-12 16:46:12 Local0.Info 192.168.100.254 Aug 12 16:46:12 pf: 54. 506486 rule 142/0(match): pass in on ste1: (tos 0x0, ttl 114, id 48361, offset 0, flags [DF], proto TCP (6), length 48) 59.182.35.69.8318 > 192.168.100.5.25: [|tcp]
2008-08-12 16:47:14 Local0.Info 192.168.100.254 Aug 12 16:47:14 pf: 62. 320660 rule 142/0(match): pass in on ste1: (tos 0x0, ttl 117, id 50815, offset 0, flags [DF], proto TCP (6), length 48) 88.156.199.178.2415 > 192.168.100.5.25: [|tcp]
2008-08-12 16:48:23 Local0.Info 192.168.100.254 Aug 12 16:48:23 pf: 69. 379247 rule 142/0(match): pass in on ste1: (tos 0x0, ttl 111, id 65277, offset 0, flags [DF], proto TCP (6), length 48) 86.110.187.14.4613 > 192.168.100.5.25: [|tcp]
2008-08-12 16:48:34 Local0.Info 192.168.100.254 Aug 12 16:48:34 pf: 10. 291989 rule 142/0(match): pass in on ste1: (tos 0x0, ttl 115, id 23925, offset 0, flags [DF], proto TCP (6), length 48) 89.169.159.198.19732 > 192.168.100.5.25: [|tcp]
2008-08-12 16:49:36 Local0.Info 192.168.100.254 Aug 12 16:49:36 pf: 62. 583301 rule 142/0(match): pass in on ste1: (tos 0x0, ttl 115, id 62019, offset 0, flags [DF], proto TCP (6), length 48) 89.169.159.198.20202 > 192.168.100.5.25: [|tcp]
2008-08-12 16:51:43 Local0.Info 192.168.100.254 Aug 12 16:51:43 pf: 126. 487669 rule 142/0(match): pass in on ste1: (tos 0x0, ttl 115, id 1366, offset 0, flags [DF], proto TCP (6), length 48) 89.169.159.198.21415 > 192.168.100.5.25: [|tcp]
2008-08-12 17:09:17 Local0.Info 192.168.100.254 Aug 12 17:09:16 pf: 1053. 663179 rule 142/0(match): pass in on ste1: (tos 0x0, ttl 105, id 16896, offset 0, flags [none], proto TCP (6), length 44) 116.41.71.126.55633 > 192.168.100.5.25: [|tcp]
2008-08-12 18:08:57 Daemon.Info 192.168.100.254 Aug 12 18:08:51 dhclient[275]: DHCPDISCOVER on ste0 to 255.255.255.255 port 67 interval 2
2008-08-12 18:08:57 Auth.Info 192.168.100.254 Aug 12 18:08:51 sshd[300]: Server listening on :: port 22.
2008-08-12 18:08:57 Auth.Info 192.168.100.254 Aug 12 18:08:51 sshd[300]: Server listening on 0.0.0.0 port 22.
2008-08-12 18:08:57 Auth.Error 192.168.100.254 Aug 12 18:08:51 sshlockout[302]: sshlockout starting up
2008-08-12 18:08:58 Daemon.Info 192.168.100.254 Aug 12 18:08:52 dhclient[275]: DHCPDISCOVER on ste0 to 255.255.255.255 port 67 interval 3

My current working theory is that there is a process that goes CPU bound and I stops me getting back in but then on a reboot I would expect the CPU temp to be a lot higher than 42 degrees after running at %100 for over an hour.

Cheers

sullrich

I am thinking this might be a driver issue with ste on RELENG_7_0 FreeBSD.

cheesyboofs

:o Noooooooooo! :'( I can't write my router off just yet, I've only just finished building it.

Joke's aside, if 'patches' are released for FreeBSD 7 will the latest ones be incorporated in to 1.2.1-Release? or is the version of 7 you start with the version of 7 you go to press with? i.e. will the version of freeBSD 7 that comes with 1.2.1 be the same as the one that comes with 1.3.

Regards

EDIT: PS, would the D-link driver going AWOL also lock-up my on-board 1Gb Realtek?