New 5400 box goes into a reboot loop after rebuild to R80.20, any ideas?

[Dario1]

Hello gents, any info or possible resolution for this issue would be much appreciated.

We have had a box RMAd due to PSU failure, the new box has R80.40 preinstalled however we need to downgrade OS to match our other GW in the HA cluster to R80.20. I have manually installed off-line CPUSE image which installs successfully, upon reboot it goes into reboot loop. then I downloaded blink image and installed following the blink instructions also installs successfully then same again reboot goes into endless loop. To restore box back to original R80.40 i would interrupt the boot and restore the snapshot.

I have never had any issues with the builds using either method in the past the only difference here is this box came with R80.40 preinstalled. Most of our builds are done remotely with terminal server console, any ideas where we going wrong?

Here is the output post build and reboot: (Appreciate your help) 

Tue May 4 11:07:57 2021 *A* [Main]: Blink image execution started
Tue May 4 11:07:57 2021 *A* [Main]: Blink Utility version: 1.1
Tue May 4 11:07:57 2021 *N* [Main]: Extracting image content to: /var/log/blink/launcher/files/
Tue May 4 11:08:03 2021 *N* [Main]: Verifying main engine integrity
Tue May 4 11:08:03 2021 *A* [Main]: Verified OK
Tue May 4 11:08:03 2021 *N* [Main]: Loading engine configurations
Tue May 4 11:08:03 2021 *A* [Main]: Image Version: R80.20
Tue May 4 11:08:03 2021 *A* [Main]: Blink Engine Version: 1.1
Tue May 4 11:08:03 2021 *N* [Main]: Image Used: blink_image_1.1_Check_Point_R80.20_T117_Gateway.tgz
Tue May 4 11:08:03 2021 *N* [Main]: Executing the main engine
Tue May 4 11:08:04 2021 *A* [Main]: Starting Blink engine process
Tue May 4 11:08:04 2021 *N* [Main]: Preparing required installation files
Tue May 4 11:08:04 2021 *N* [Main]: Verifying the integrity of the image files
Tue May 4 11:08:13 2021 *A* [Main]: Verification succeeded
Tue May 4 11:08:13 2021 *A* [Main]: Starting installation of 'Security Gateway' image
Tue May 4 11:08:13 2021 *N* [Main]: Executing stage - Create Partition
Tue May 4 11:08:18 2021 *N* [Main]: Executing stage - Extract Image
Tue May 4 11:08:56 2021 *N* [Main]: Executing stage - Merge /var/log files
Tue May 4 11:08:57 2021 *N* [Main]: Executing stage - Create Snapshot
Tue May 4 11:08:57 2021 *N* [Main]: Skipping 'Create Snapshot' stage.
Tue May 4 11:08:57 2021 *N* [Main]: Executing stage - Machine Configuration
Tue May 4 11:09:04 2021 *N* [Main]: Executing stage - Gateway Configuration
Tue May 4 11:09:04 2021 *N* [Main]: Executing stage - Blink Wizard Configurations
Tue May 4 11:09:04 2021 *N* [Main]: Skipping 'Blink Wizard Configurations' stage.
Tue May 4 11:09:04 2021 *N* [Main]: Executing stage - Blink Updates Configurations
Tue May 4 11:09:04 2021 *N* [Main]: Skipping 'Blink Updates Configurations' stage.
Tue May 4 11:09:04 2021 *N* [Main]: Executing stage - User Updates Configurations
Tue May 4 11:09:04 2021 *N* [Main]: Skipping 'User Updates Configurations' stage.
Tue May 4 11:09:04 2021 *N* [Main]: Executing stage - Post Installation Actions
Tue May 4 11:09:19 2021 *N* [Main]: Executing stage - Finalize Installation
Tue May 4 11:09:19 2021 *N* [Main]: Total time for gateway 'Security Gateway': 0 hours 1 minutes 6 seconds
Tue May 4 11:09:19 2021 *A* [Main]: Security Gateway execution finished
Tue May 4 11:09:19 2021 *N* [Main]: Total Blink engine execution time: 0 hours 1 minutes 15 seconds
Tue May 4 11:09:19 2021 *N* [Main]: The machine will go to reboot in 10 seconds!
Tue May 4 11:09:19 2021 *A* [Main]: The installation has finished successfully and is pending reboot!
[Expert@gw-838e5f:0]#
Broadcast message from admin@gw-838e5f (Tue May 4 11:09:29 2021):

Please stand by while rebooting the system...

ACPI: Getting cpuindex for acpiid 0x3
ACPI: Getting cpuindex for acpiid 0x4
ACPI: Getting cpuindex for acpiid 0x5
ACPI: Getting cpuindex for acpiid 0x6ooting in 4 seconds]
ACPI: Getting cpuindex for acpiid 0x7
ACPI: Getting cpuindex for acpiid 0x8
ÿi8042.c: No controller found.
insmod: error inserting '/lib/cciss.ko': -1 Unknown symbol in module
insmod: error inserting '/lib/ahci.ko': -1 Unknown symbol in module
read_urandom: /dev/urandom: open failed: No such file or directory
Reading all physical volumes. This may take a while...
/dev/ram0: ioctl BLKPBSZGET failed: Invalid argumentds]
/dev/ram0: ioctl BLKPBSZGET failed: Invalid argument
/dev/ram0: ioctl BLKPBSZGET failed: Invalid argument
/dev/ram0: ioctl BLKPBSZGET failed: Invalid argument

....................... more of the same after this

/dev/ram7: ioctl BLKPBSZGET failed: Invalid arguKernel panic - not syncing: Attempted to kill init!
ment
/dev/ram 7: ioctl BLKPBSZ
Entering kdb (current=0xffff810107f447b0, pid 1) on processor 0 due to KDB_ENTER()
ahci: Unknown symbol ata_link_online
ahci: Unknown symbol ata_std_postreset
ahci: Unknown symbol sata_pmp_std_postreset
ahci: Unknown symbol sata_pmp_std_prereset
ahci: Unknown symbol ata_tf_from_fis
ahci: Unknown symbol ata_qc_complete_multiple
ahci: Unknown symbol ata_wait_after_reset
ahci: Unknown symbol ata_wait_ready
ahci: Unknown symbol ata_noop_dev_select
ahci: Unknown symbol ata_tf_to_fis
ahci: Unknown symbol ata_pci_device_do_resume
ahci: Unknown symbol ata_host_activate
ahci: Unknown symbol ata_dev_classify
ahci: Unknown symbol sata_link_hardreset
ahci: Unknown symbol ata_scsi_change_queue_depth
ahci: Unknown symbol sata_std_hardreset
ahci: Unknown symbol class_device_attr_link_power_management_policy
ahci: Unknown symbol ata_ehi_clear_desc
ahci: Unknown symbol ata_pci_device_suspend

....................... more of the same after this

ioctl32(lvm:609): Unknown cmd fd(3) cmd(0000127b){00} arg(0933e0e8) on /dev/ram8
usb 1-1: new high speed USB device using ehci_hcd and address 2
usb 1-1: configuration #1 chosen from 1 choice
hub 1-1:1.0: USB hub found
hub 1-1:1.0: 6 ports detected
usb 2-1: new high speed USB device using ehci_hcd and address 2
usb 2-1: configuration #1 chosen from 1 choice
hub 2-1:1.0: USB hub found
hub 2-1:1.0: 8 ports detected
Kernel panic - not syncing: Attempted to kill init!
wdt stop function not defined
Stack traceback for pid 1
0xffff810107f447b0 1 0 1 0 R 0xffff810107f44a90 *init
rsp rip Function (args)
0xffff810107f57e00 0xffffffff8015cf9c kdb_panic+0x1c
0xffff810107f57e00 0xffffffff80303227 notifier_call_chain+0x37 (unknown, unknown, unknown)
0xffff810107f57e20 0xffffffff80040fa9 panic+0xa9 (0xffffffff803a7e42, 0x292, 0x0, 0x1, 0x3, 0x1)
0xffff810107f57f10 0xffffffff800450f5 do_exit+0x975 (0x100)
0xffff810107f57f60 0xffffffff8004521e do_group_exit+0x4e (unknown)
0xffff810107f57f80 0xffffffff8002d253 sysenter_do_call+0x1b (unknown, unknown, unknown, unknown, unknown, unknown, 0x23, 0x296)

ACPI: Getting cpuindex for acpiid 0x3
ACPI: Getting cpuindex for acpiid 0x4ooting in 4 seconds]
ACPI: Getting cpuindex for acpiid 0x5
ACPI: Getting cpuindex for acpiid 0x6
ACPI: Getting cpuindex for acpiid 0x7
ACPI: Getting cpuindex for acpiid 0x8
ÿi8042.c: No controller found.
insmod: error inserting '/lib/cciss.ko': -1 Unknown symbol in module
insmod: error inserting '/lib/ahci.ko': -1 Unknown symbol in module
read_urandom: /dev/urandom: open failed: No such file or directory
Reading all physical volumes. This may take a while...
/dev/ram0: ioctl BLKPBSZGET failed: Invalid argument
/dev/ram0: ioctl BLKPBSZGET failed: Invalid argument
/dev/ram0: ioctl BLKPBSZGET failed: Invalid argument

Then does this over and over again.

Any ideas? Tx

[PhoneBoy]

Recommend a TAC case.

[shlomip]

Hi @Dario1 , 

You are trying to downgrade R80.40 ( 3.10 kernel)  to R80.20 (2.6.18 kernel) this is not supported. (different file system)

See the below references:

- R80.40 Known limitation 

- sk170954

Best is to do Clean install of your R80.20 image on the RMA'ed device

[Dario1]

Thanks PhoneBoy, TAC was my next point of call. I tend to have a look and post on this and previous forums as this forum has helped me to resolve issues quick countless times. Appreciate your hard work boys.

Shlomip, you right of course the new box is XFS instead of EXT3 file system, I did do what i thought was a clean install "installer clean-install" which does successfully install but fails to setup the boot variables in some way. So is it the case that the newly build R80.20 partition is EXT3 but the boot part of the device stays XFS hence failing the boot sequence as it cant read the boot files relating to EXT3? 

I would have thought fair few people would have had this issue, wonder if there is a way of rolling back to EXT3 as part of clean-install/rebuild? That sk relates to restoring the snapshot where as I am rebuilding the OS then restoring the backup, so very close but not the same. There must be a way.

Many Thanks. D

installer import local /var/log/r8020_upgrade_files/Check_Point_R80.20_T101_Fresh_Install_and_Upgrade_Security_Gateway.tgz

installer clean-install

[Expert@gw-838e5f:0]# df -T
Filesystem Type 1K-blocks Used Available Use% Mounted on
/dev/mapper/vg_splat-lv_current xfs 33538048 7497628 26040420 23% /

[PhoneBoy]

Installing from ISO is the only way to achieve this.
This means booting the appliance from a USB drive prepared with ISOmorphic using the appropriate installation image. 
This completely reformats/repartitions the drive.

[Dario1]

I was afraid you would say that, I always thought that mounting a local ISO image as a USB drive is very similar if not the same as using ISOmorphic USB build disk. This is a problem for us as alot of sites are unmanned and the kit is racked with a standard console and layer 3 interface which used to allow remote build where as now its a trip to a remote site and significant delay to restoring RMAd kit.

Thanks for the update. D

[PhoneBoy]

It's not the same precisely because the disk itself needs to be reformatted and repartitioned, which can only be done when you are not booting from said disk.
I believe you would be able to achieve this remotely with a LOM.