Proventia IBM Gx3002 install guidance
-
I assume you're using a serial console, yes?
Yes I am using the serial console with a null modem cable.
When you installed to the HD in a different machine did you select the embedded kernel or enable serial console in the webgui?
I tried both, at least I thought I had. My experience this morning tells me I didn't.
-
More as a reference for other folks…
To disable the bypass:
Remove the two screws on each side and pop the covers apart.With the board up and the e'net ports facing you, you'll see two red jumpers near the edge facing you. The "top" jumper aka the one furthest from you, is the the one you want to move to the right. The jumper below is probably already on the rightmost two pins.
At this point you may want to remove the PATA hard drive because of what I'll suggest next. I also removed the frame that holds the drive to get better airflow to the components.
Re-assemble.
Flip the unit so the bottom is facing you. There's a plate to be removed. Two of the screws show and 3 are hidden by the mfg's sticker (most likely). If you remove the plate you'll find the so-dimm socket for the system memory AND a CF socket that is on the second PATA controller as the primary. I burned the embedded 2 gig nanobsd (serial console) onto a 2gig CF, and installed it here.
Once you re-assemble, the only thing that moves is the cpu fan.
-
Im also in the process of installing one of these boxes. First tried the USB drive imge: pfSense-memstick-2.1-RELEASE-i386.img.gz
Nothing continued to boot after it got past the bios checks.
Phoenix - AwardBIOS v6.00PG, An Energy Star Ally
Copyright (C) 1984-2003, Phoenix Technologies, LTDProventia BIOS GX3002 Version R1.00.W1 T8 ( 11212006 )
Main Processor : VIA C7 1.00GHz(100x10.0)
Memory Testing : 1039296K OK + 8M shared memoryDDR DIMM Speed : 533
DDR DIMM Data Width : 64-Bit, Single Channel
IDE Channel 0 Master : None
IDE Channel 0 Slave : None
IDE Channel 1 Master : None
IDE Channel 1 Slave : None
Phoenix Technologies, LTD
System Configurations
+==============================================================================+
| CPU Type : VIA C7 Base Memory : 640K |
| CPU ID/ucode ID : 06A9 Extended Memory :1038336K |
| CPU Clock : 1.00GHz Cache Memory : 128K |
|–----------------------------------------------------------------------------|
| Diskette Drive A : None Display Type : EGA/VGA |
| Diskette Drive B : None Serial Port(s) : 3F8 2F8 |
| Pri. Master Disk : None Parallel Port(s) : None |
| Pri. Slave Disk : None DDR2 SDRAM at Rank : 0 1 |
| Sec. Master Disk : None |
| Sec. Slave Disk : None |
+==============================================================================+USB Storage Device : VBTM Store 'n' Go 5.00...Full Speed device 1.X
PCI device listing ...
Bus No. Device No. Func No. Vendor/Device Class Device Class IRQ0 8 0 10EC 8139 0200 Network Cntrlr 11
0 9 0 10EC 8139 0200 Network Cntrlr 7
0 10 0 10EC 8139 0200 Network Cntrlr 5
0 15 0 1106 0571 0101 IDE Cntrlrtrlr 14
0 16 0 1106 3038 0C03 USB 1.0/1.1 UHCI Cntrlr 11
0 16 1 1106 3038 0C03 USB 1.0/1.1 UHCI Cntrlr 11
0 16 2 1106 3038 0C03 USB 1.0/1.1 UHCI Cntrlr 7
0 16 3 1106 3038 0C03 USB 1.0/1.1 UHCI Cntrlr 7
0 16 4 1106 3104 0C03 USB 2.0 EHCI Cntrlr 5
1 0 0 1106 3344 0300 Display Cntrlr 11 -
And here is the post from when trying to use the CF card with pfSense-2.1-RELEASE-4g-i386-nanobsd.img.gz
Phoenix - AwardBIOS v6.00PG, An Energy Star Ally
Copyright (C) 1984-2003, Phoenix Technologies, LTDProventia BIOS GX3002 Version R1.00.W1 T8 ( 11212006 )
Main Processor : VIA C7 1.00GHz(100x10.0)
Memory Testing : 1039296K OK + 8M shared memoryDDR DIMM Speed : 533
DDR DIMM Data Width : 64-Bit, Single Channel
IDE Channel 0 Master : None
IDE Channel 0 Slave : None
IDE Channel 1 Master : SanDisk SDCFH-004G HDX 6.03
IDE Channel 1 Slave : None
Phoenix Technologies, LTD
System Configurations
+==============================================================================+
| CPU Type : VIA C7 Base Memory : 640K |
| CPU ID/ucode ID : 06A9 Extended Memory :1038336K |
| CPU Clock : 1.00GHz Cache Memory : 128K |
|–----------------------------------------------------------------------------|
| Diskette Drive A : None Display Type : EGA/VGA |
| Diskette Drive B : None Serial Port(s) : 3F8 2F8 |
| Pri. Master Disk : None Parallel Port(s) : None |
| Pri. Slave Disk : None DDR2 SDRAM at Rank : 0 1 |
| Sec. Master Disk : LBA,ATA 33, 4000MB |
| Sec. Slave Disk : None |
+==============================================================================+PCI device listing ...
Bus No. Device No. Func No. Vendor/Device Class Device Class IRQ0 8 0 10EC 8139 0200 Network Cntrlr 11
0 9 0 10EC 8139 0200 Network Cntrlr 7
0 10 0 10EC 8139 0200 Network Cntrlr 5
0 11 0 10EC 8139 0200 Network Cntrlr 10
0 16 0 1106 3038 0C03 USB 1.0/1.1 UHCI Cntrlr 11
0 16 1 1106 3038 0C03 USB 1.0/1.1 UHCI Cntrlr 11
0 16 2 1106 3038 0C03 USB 1.0/1.1 UHCI Cntrlr 7
0 16 3 1106 3038 0C03 USB 1.0/1.1 UHCI Cntrlr 7
0 16 4 1106 3104 0C03 USB 2.0 EHCI Cntrlr 5
1 0 0 1106 3344 0300 Display Cntrlr 11 -
Apparently, Mac has a different way of writing the image than physdiskwrite on windows. Booting just fine now.
-
Anyone know where more of these can be picked up from?
-
You just have to get lucky on Ebay. ;)
Steve
-
Yeah kicking myself when i saw a auction with quantity 50 and I only got 2 about a year ago. Haven't seen any sense.
BTW has anyone tried the other proventia hardware? The GX4004-C seems to be available and is already rackmount ready. I can't seem to find information on the board itself to see if it's running the correct processor to even attempt an install.
-
I did the same…only bought two.
However I just bought another model E1500. They are rackmount, P4 3.x Ghz, 2GB RAM and 80GB Hard Drive. They have (5) Gigabit ports and one PCI slot. So hardware sounds great but I'm having an issue with the ports. for some reason I can't get them to work or only 1 or 2 (very odd). Its seems there could be linked ports like on the Gx3002 but can't find any jumper(s) or setting to change. Anyone work with one of these before and have any ideas? If someone can help me get all the ports going I would be happy to send them one of the units as a thank you.
Thanks,
Adam -
You mean the ES1500? This?
Can you access the BIOS? If this box is using LAN by-pass it's often configurable there.
Steve
-
Steve,
Thank you for the reply. Yes, the ES1500. I have checked in the BIOS but did not find anything there.
Thanks,
Adam -
Ok, so after further work it seems that they are not actually by-pass. It seems there might be driver or config issues. I brought all the connection down from the shell and then brought one by one up manually. After that each seem to work. I then enabled one by one in pfsense and confirmed each working. Rebooted a few times to make sure they kept working and no issue. Not sure why this is because I was even getting pfsense reporting the interface link up even with no cable plugged in before I went through these steps. It seems there is some others that have had similar issues with watchguard devices which use the same interface chipset. I did find the below link about updating the firmware but have not tried that and not sure it is needed.
https://forum.pfsense.org/index.php/topic,20095.msg250430/topicseen.html#msg250430
So at this time I have everything working great and will continue testing other the next several days to see if there is any stability issues.
-
Ah ok.
The Marvell 88e8053 NICs as reported in the link can suffer a lock-up. In the Watchguard boxes it usually only happens under heavy load and is reported as a "watchdog timeout" in the logs. There is an easy workaround though:
https://doc.pfsense.org/index.php/PfSense_on_Watchguard_Firebox#Known_Issues_2
The firmware (actually more like a group of register settings) update didn't help the watchdog issue. It only removed errors from yhd boot log. Are you seeing those errors? Which version are you NICs running? What does the following report:pciconf -l | grep msk
Steve