Pfsense update dont work on Alix board
-
First, thank you all for your help and messages.
jimp
I think I have gave a bad title to my thread, I dont want to say that pfSense dont work on all Alix, but it dont work good on my actual setup.
I really appreciate the work of all people that making this very very good firwall, and all this free. The goal here is to explain my problem, find the solution and, perhaps, improve a future release.It's good to know you are doing test on Alix platform. I was not sure because there are a lot hardware that pf can run on…
With compact flash on Alix, do you remember about the time that the update took ?igor
Thank you for your answer. I think as you : the compact flash is my problem. But I am not sure if it is only hardware, or the way of software is using it; perhaps is not very efficient..
I have done some tests SCP'ing update to my Alix with compact flash :
In /tmp/ it is coping at about 1500 kB/s (but unfortunately there is not enough space here...)
On /root/ it is copying at about 38 kB/s...Why this huge difference ?
Otherwise, I already have the bios 0.99h on my board. I have tested with two compact flashs (both manufactured in 2008) and on the other it is ever worst.. copying as low as 15kB/s....
Think I will buy a newer 4GB card.marcelloc
Good idea, I have tried this option, without success... after uploading compete (that took a lot of time), the update is done quickly, but when rebooting pfSense is still in 2.0 release.
I can post upgrade log if someone can exactract usefull information from it...Last precision, all this was done on fresh install of pfSense 2.0 : no package, only configuration was changing LAN address and enabling SSH.
Is anybody has an idea, I can do more tests.
Thank's all ;)
-
/tmp is a RAM drive, /root is on CF
-
Thank you.
I have done other tests :
Old Pentium 4 and one of my compact flash on IDE : same problem
Very old Pentium 3 and with the same compact flash : same problemAs I can see, the problem is my compact flash cards.
On the ram drive, I always have honorable speed (1.5-4 Mo/s, depending configuration) but on compact flash it is always in the range of 10-40 ko/s.
I have benchmarked compact flash on Windows, and with medium/large file (larger then 8ko) I can write at 4 Mo/s
My topic title is wrong, but I can't edit it.
What should I do ? Open a new one? It is normal that on freebsd/pfsense I have very low write speed and with the same card a relatively correct transfert on Windows ?Thanks for you help & ideas ;)
-
Hi ;)
It seems that disk access speed does not interest many people….
I have bought a new 4Go compact flash at PC Engines, and now update works ! I have reinstalled the old 2.0 and then successfully upgraded to 2.01 via Web GUI.
igor , you were 100% right, thanks ;)
But the question remains : I have benchmarked the new compact flash at more than 15 MB/s in write, but the upload to pfsense is still at about 500kB/s....
I dont know if it is normal, or if it can be improved.Thanks all for your help :)
-
Take a look on this topic, however I don't know how to implement this on compact flash :(
http://forum.pfsense.org/index.php/topic,43737.0.html
-
That won't really help CF anyhow. The problem is that the ALIX boards (possibly just with FreeBSD?) don't have proper DMA, so they will be slower even still on top of CF being slow in general.
Windows is probably using DMA when talking to the card, so that's why it's faster.
Check your boot messages, you'll probably see it like so:
ad0: 3815MB <sandisk sdcfh-004g="" hdx="" 6.03="">at ata0-master PIO4</sandisk> -
Yes, it is possible, alix boards use very old components (AMD Geode & companion).
I'll check the next time I'll reboot it if it use PIO instead DMA.Thanks ;)
-
I auto upgrade PFSense 2.0 to 2.01 successfully. Depending on the number of add-on packages you previously installed before the upgrade, it might take a bit longer to complete. In my case I had to leave it for half a day.
Hardware:
Alix 2D13 with micro drive.
-
Hi to everyboby. I can't upgrade on ALIX board from 2.0 to 2.0.1 in any way (web interface, command line from URL or local).
Someone can help me? This is the upgrade log… maybe an hardware problem to the CF?NanoBSD Firmware upgrade in progress…
Installing /root/latest-nanobsd-4g.img.gz.
SLICE 2
OLDSLICE 1
TOFLASH ad0s2
COMPLETE_PATH ad0s2a
GLABEL_SLICE pfsense1
Thu Mar 8 10:04:31 CET 2012total 4
dr-xr-xr-x 8 root wheel 512B Mar 8 10:02 .
drwxr-xr-x 25 root wheel 512B Mar 8 10:02 ..
crw-r----- 1 root operator 0, 56 Mar 8 10:02 ad0
crw-r----- 1 root operator 0, 57 Mar 8 10:02 ad0s1
crw-r----- 1 root operator 0, 60 Mar 8 10:02 ad0s1a
crw-r----- 1 root operator 0, 58 Mar 8 10:02 ad0s2
crw-r----- 1 root operator 0, 61 Mar 8 10:02 ad0s2a
crw-r----- 1 root operator 0, 59 Mar 8 10:02 ad0s3
crw------- 1 root operator 0, 28 Mar 8 10:02 ata
crw------- 1 root wheel 0, 11 Mar 8 10:04 bpf
lrwxr-xr-x 1 root wheel 3B Mar 8 10:02 bpf0 -> bpf
crw------- 1 root tty 0, 5 Mar 8 10:04 console
crw-rw-rw- 1 root wheel 0, 44 Mar 8 10:02 crypto
crw-rw-rw- 1 root wheel 0, 10 Mar 8 10:02 ctty
crw-rw---- 1 uucp dialer 0, 35 Mar 8 10:02 cuau0
crw-rw---- 1 uucp dialer 0, 36 Mar 8 10:02 cuau0.init
crw-rw---- 1 uucp dialer 0, 37 Mar 8 10:02 cuau0.lock
crw-rw---- 1 uucp dialer 0, 41 Mar 8 10:02 cuau1
crw-rw---- 1 uucp dialer 0, 42 Mar 8 10:02 cuau1.init
crw-rw---- 1 uucp dialer 0, 43 Mar 8 10:02 cuau1.lock
crw------- 1 root wheel 0, 4 Mar 8 10:02 devctl
cr-------- 1 root wheel 0, 55 Mar 8 10:02 devstat
dr-xr-xr-x 2 root wheel 512B Mar 8 10:02 fd
crw------- 1 root wheel 0, 13 Mar 8 10:02 fido
crw-r----- 1 root operator 0, 3 Mar 8 10:02 geom.ctl
crw------- 1 root wheel 0, 23 Mar 8 10:02 io
crw------- 1 root wheel 0, 6 Mar 8 10:02 klog
crw-r----- 1 root kmem 0, 15 Mar 8 10:02 kmem
dr-xr-xr-x 2 root wheel 512B Mar 8 10:02 led
crw-r----- 1 root operator 0, 70 Mar 8 10:02 md0
crw-r----- 1 root operator 0, 71 Mar 8 10:02 md1
crw------- 1 root wheel 0, 48 Mar 8 10:02 mdctl
crw-r----- 1 root kmem 0, 14 Mar 8 10:02 mem
crw------- 1 root kmem 0, 16 Mar 8 10:02 nfslock
crw-rw-rw- 1 root wheel 0, 25 Mar 8 10:04 null
crw-r--r-- 1 root wheel 0, 27 Mar 8 10:02 pci
crw-rw---- 1 root proxy 0, 45 Mar 8 10:02 pf
crw-rw-rw- 1 root wheel 0, 9 Mar 8 10:02 ptmx
dr-xr-xr-x 2 root wheel 512B Mar 8 10:03 pts
crw-rw-rw- 1 root wheel 0, 7 Mar 8 10:02 random
crw------- 1 root wheel 0, 24 Mar 8 10:02 speaker
lrwxr-xr-x 1 root wheel 4B Mar 8 10:02 stderr -> fd/2
lrwxr-xr-x 1 root wheel 4B Mar 8 10:02 stdin -> fd/0
lrwxr-xr-x 1 root wheel 4B Mar 8 10:02 stdout -> fd/1
crw------- 1 root wheel 0, 32 Mar 8 10:02 ttyu0
crw------- 1 root wheel 0, 33 Mar 8 10:02 ttyu0.init
crw------- 1 root wheel 0, 34 Mar 8 10:02 ttyu0.lock
crw------- 1 root wheel 0, 38 Mar 8 10:02 ttyu1
crw------- 1 root wheel 0, 39 Mar 8 10:02 ttyu1.init
crw------- 1 root wheel 0, 40 Mar 8 10:02 ttyu1.lock
crw------- 1 uucp dialer 0, 62 Mar 8 10:03 tun1
dr-xr-xr-x 2 root wheel 512B Mar 8 10:02 ufs
dr-xr-xr-x 2 root wheel 512B Mar 8 10:02 ufsid
lrwxr-xr-x 1 root wheel 9B Mar 8 10:02 ugen0.1 -> usb/0.1.0
lrwxr-xr-x 1 root wheel 9B Mar 8 10:02 ugen1.1 -> usb/1.1.0
lrwxr-xr-x 1 root wheel 6B Mar 8 10:02 urandom -> random
dr-xr-xr-x 2 root wheel 512B Mar 8 10:02 usb
crw-r--r-- 1 root operator 0, 46 Mar 8 10:02 usbctl
crw------- 1 root operator 0, 47 Mar 8 10:02 xpt0
crw-rw-rw- 1 root wheel 0, 26 Mar 8 10:02 zero-rw-r--r-- 1 root wheel 67M Mar 8 09:50 /root/latest-nanobsd-4g.img.gz
MD5 (/root/latest-nanobsd-4g.img.gz) = 0fc0188215421c7eaf978fa14a60c7db
/dev/ufs/pfsense0 on / (ufs, local, noatime, synchronous)
devfs on /dev (devfs, local)
/dev/md0 on /tmp (ufs, local)
/dev/md1 on /var (ufs, local)
devfs on /var/dhcpd/dev (devfs, local)last pid: 45360; load averages: 2.97, 1.10, 0.43 up 0+00:02:26 10:04:35
46 processes: 2 running, 44 sleepingMem: 53M Active, 93M Inact, 33M Wired, 8K Cache, 34M Buf, 55M Free
Swap:PID USERNAME THR PRI NICE SIZE RES STATE TIME WCPU COMMAND
30795 root 1 116 20 38236K 19768K RUN 0:01 1.95% php
51405 root 1 76 0 34140K 19492K accept 0:01 0.00% php
18135 root 1 76 0 36188K 16536K nanslp 0:01 0.00% php
19199 root 1 44 0 7992K 3520K select 0:00 0.00% sshd
31515 root 1 76 0 7992K 3520K select 0:00 0.00% sshd
46432 root 1 76 0 32092K 9960K wait 0:00 0.00% php
45963 root 1 76 0 32092K 9904K wait 0:00 0.00% php
17129 root 1 44 0 5988K 3928K select 0:00 0.00% racoon
19049 root 1 44 0 7208K 5972K select 0:00 0.00% bsnmpd
26586 root 1 44 0 4944K 2392K select 0:00 0.00% syslogd
54860 nobody 1 44 0 5556K 2516K select 0:00 0.00% dnsmasq
45575 root 1 44 0 5564K 3820K kqread 0:00 0.00% lighttpd
32771 root 1 76 20 3656K 1392K wait 0:00 0.00% sh
27000 root 1 44 0 5912K 2276K bpf 0:00 0.00% tcpdump
27105 root 1 44 0 3316K 872K piperd 0:00 0.00% logger
35140 root 1 76 0 3656K 1476K wait 0:00 0.00% sh
52701 root 1 76 0 3684K 1596K wait 0:00 0.00% login
40035 root 1 44 0 3316K 1296K select 0:00 0.00% apingerNanoBSD upgrade starting
dd if=/dev/zero of=/dev/ad0s2 bs=1m count=1
1+0 records in
1+0 records out
1048576 bytes transferred in 0.995684 secs (1053121 bytes/sec)/usr/bin/gzip -dc /root/latest-nanobsd-4g.img.gz | /bin/dd of=/dev/ad0s2 obs=64k
dd: /dev/ad0s2: short write on character device
dd: /dev/ad0s2: end of device
3510386+947 records in
27428+1 records out
1797530112 bytes transferred in 434.345388 secs (4138481 bytes/sec)
After upgrade fdisk/bsdlabel/sbin/fsck_ufs -y /dev/ad0s2a
fsck_ufs: /dev/ad0s2a: can't read disk label
** /dev/ad0s2aCANNOT READ BLK: 3836192
CONTINUE? yesTHE FOLLOWING DISK SECTORS COULD NOT BE READ: 3836192, 3836193, 3836194,
ioctl (GCINFO): Inappropriate ioctl for deviceFile list:
Warning: file_get_contents(/conf/file_upgrade_log.txt): failed to open stream: No such file or directory in /usr/local/www/diag_nanobsd.php on line 236
Misc log:
Warning: file_get_contents(/conf/firmware_update_misc.log): failed to open stream: No such file or directory in /usr/local/www/diag_nanobsd.php on line 238
fdisk/bsdlabel log:
Before upgrade fdisk/bsdlabel
******* Working on device /dev/ad0 *******
parameters extracted from in-core disklabel are:
cylinders=7314 heads=16 sectors/track=63 (1008 blks/cyl)Figures below won't work with BIOS for partitions not in cyl 1
parameters to be used for BIOS calculations are:
cylinders=7314 heads=16 sectors/track=63 (1008 blks/cyl)Media sector size is 512
Warning: BIOS sector numbering starts with sector 1
Information from DOS bootblock is:
The data for partition 1 is:
sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD)
start 63, size 3861585 (1885 Meg), flag 80 (active)
beg: cyl 0/ head 1/ sector 1;
end: cyl 758/ head 15/ sector 63
The data for partition 2 is:
sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD)
start 3861711, size 3861585 (1885 Meg), flag 0
beg: cyl 759/ head 1/ sector 1;
end: cyl 493/ head 15/ sector 63
The data for partition 3 is:
sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD)
start 7723296, size 102816 (50 Meg), flag 0
beg: cyl 494/ head 0/ sector 1;
end: cyl 595/ head 15/ sector 63
The data for partition 4 is:
<unused># /dev/ad0s1:
type: unknown
disk: amnesiac
label:
flags:
bytes/sector: 512
sectors/track: 63
tracks/cylinder: 16
sectors/cylinder: 1008
cylinders: 3830
sectors/unit: 3861585
rpm: 3600
interleave: 1
trackskew: 0
cylinderskew: 0
headswitch: 0 # milliseconds
track-to-track seek: 0 # milliseconds
drivedata: 08 partitions:
# size offset fstype [fsize bsize bps/cpg]
a: 3861569 16 unused 0 0
c: 3861585 0 unused 0 0 # "raw" part, don't edit/dev/ad0s2:
type: unknown
disk: amnesiac
label:
flags:
bytes/sector: 512
sectors/track: 63
tracks/cylinder: 16
sectors/cylinder: 1008
cylinders: 3823
sectors/unit: 3854529
rpm: 3600
interleave: 1
trackskew: 0
cylinderskew: 0
headswitch: 0 # milliseconds
track-to-track seek: 0 # milliseconds
drivedata: 08 partitions:
# size offset fstype [fsize bsize bps/cpg]
a: 3854513 16 unused 0 0
c: 3854529 0 unused 0 0 # "raw" part, don't edit
–-------------------------------------------------------------******* Working on device /dev/ad0 *******
parameters extracted from in-core disklabel are:
cylinders=7314 heads=16 sectors/track=63 (1008 blks/cyl)Figures below won't work with BIOS for partitions not in cyl 1
parameters to be used for BIOS calculations are:
cylinders=7314 heads=16 sectors/track=63 (1008 blks/cyl)Media sector size is 512
Warning: BIOS sector numbering starts with sector 1
Information from DOS bootblock is:
The data for partition 1 is:
sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD)
start 63, size 3861585 (1885 Meg), flag 80 (active)
beg: cyl 0/ head 1/ sector 1;
end: cyl 758/ head 15/ sector 63
The data for partition 2 is:
sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD)
start 3861711, size 3861585 (1885 Meg), flag 0
beg: cyl 759/ head 1/ sector 1;
end: cyl 493/ head 15/ sector 63
The data for partition 3 is:
sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD)
start 7723296, size 102816 (50 Meg), flag 0
beg: cyl 494/ head 0/ sector 1;
end: cyl 595/ head 15/ sector 63
The data for partition 4 is:
<unused># /dev/ad0s1:
type: unknown
disk: amnesiac
label:
flags:
bytes/sector: 512
sectors/track: 63
tracks/cylinder: 16
sectors/cylinder: 1008
cylinders: 3830
sectors/unit: 3861585
rpm: 3600
interleave: 1
trackskew: 0
cylinderskew: 0
headswitch: 0 # milliseconds
track-to-track seek: 0 # milliseconds
drivedata: 08 partitions:
# size offset fstype [fsize bsize bps/cpg]
a: 3861569 16 unused 0 0
c: 3861585 0 unused 0 0 # "raw" part, don't edit/dev/ad0s2:
type: unknown
disk: amnesiac
label:
flags:
bytes/sector: 512
sectors/track: 63
tracks/cylinder: 16
sectors/cylinder: 1008
cylinders: 3823
sectors/unit: 3854529
rpm: 3600
interleave: 1
trackskew: 0
cylinderskew: 0
headswitch: 0 # milliseconds
track-to-track seek: 0 # milliseconds
drivedata: 08 partitions:
# size offset fstype [fsize bsize bps/cpg]
a: 3854513 16 unused 0 0
c: 3854529 0 unused 0 0 # "raw" part, don't edit
–-------------------------------------------------------------</unused></unused> -
1. Please start a new thread, don't hijack an old thread.
2. The fsck errors in that log would seem to indicate your CF card is failing. -
Ok… sorry for this... :'(
Thanks.
Andrea.
