Unstable boot from USB

I am experiencing the same problems Styno describes in this closed thread.

In my case I am using a 1GB Kingston DataTraveler USB pen drive. Everything was working fine for almost 2 months (with many restarts in between). Now it has become very unstable. Sometimes I can boot from it sometimes I can't. I can not reproduce the problem. The USB pen drive works fine when using it in other Linux machine.
I have checked the pen drive with badblocks using -F option and everything went fine: no bad block.
I have deleted all partitions and reinstalled everything back in the pen drive as it was before but the problem persists. I took an old hard disk, configured exactly the same as it is in the pen drive, copied all same files into it and it boots perfectly.
It seems to me that pen drive has lost some of its characteristics I can not come up with.
Here you go the log at boot time with the pen drive plugged in. There you can see some strange things when it comes to the USB part.

Code:

Jan 1 01:00:02 kernel: ip_conntrack version 2.1 (128 buckets, 1024 max) - 344 bytes per conntrack Jan 1 01:00:02 kernel: ip_conntrack_pptp version 1.9 loaded Jan 1 01:00:02 kernel: ip_nat_pptp version 1.5 loaded Jan 1 01:00:02 kernel: ip_tables: (C) 2000-2002 Netfilter core team Jan 1 01:00:02 kernel: ipt_time loading Jan 1 01:00:02 kernel: NET4: Unix domain sockets 1.0/SMP for Linux NET4.0. Jan 1 01:00:02 kernel: IPv6 v0.8 for NET4.0 Jan 1 01:00:02 kernel: IPv6 over IPv4 tunneling driver Jan 1 01:00:02 kernel: NET4: Ethernet Bridge 008 for NET4.0 Jan 1 01:00:02 kernel: 802.1Q VLAN Support v1.7 Ben Greear <greearb@candelatech.com> Jan 1 01:00:02 kernel: All bugs added by David S. Miller <davem@redhat.com> Jan 1 01:00:02 kernel: FAT: bogus logical sector size 59904 Jan 1 01:00:02 kernel: FAT: bogus logical sector size 59904 Jan 1 01:00:02 kernel: NTFS: Unable to set blocksize 512. Jan 1 01:00:02 kernel: VFS: Mounted root (squashfs filesystem) readonly. Jan 1 01:00:02 kernel: Mounted devfs on /dev Jan 1 01:00:02 kernel: Freeing unused kernel memory: 72k freed Jan 1 01:00:02 kernel: Warning: unable to open an initial console. Jan 1 01:00:02 kernel: Algorithmics/MIPS FPU Emulator v1.5 Jan 1 01:00:02 kernel: usb.c: registered new driver usbdevfs Jan 1 01:00:02 kernel: usb.c: registered new driver hub Jan 1 01:00:02 kernel: usb-ohci.c: USB OHCI at membase 0xb8004000, IRQ 2 Jan 1 01:00:02 kernel: usb-ohci.c: usb-00:04.0, PCI device 14e4:4715 Jan 1 01:00:02 kernel: usb.c: new USB bus registered, assigned bus number 1 Jan 1 01:00:02 kernel: hub.c: USB hub found Jan 1 01:00:02 kernel: hub.c: 2 ports detected Jan 1 01:00:02 kernel: SCSI subsystem driver Revision: 1.00 Jan 1 01:00:02 kernel: hub.c: new USB device 00:04.0-1, assigned address 2 Jan 1 01:00:02 kernel: 0/0x653d) is not claimed by any active driver. Jan 1 01:00:02 kernel: Initializing USB Mass Storage driver... Jan 1 01:00:02 kernel: usb.c: registered new driver usb-storage Jan 1 01:00:02 kernel: USB Mass Storage support registered. Jan 1 01:00:03 kernel: usb.c: deregistering driver usb-storage Jan 1 01:00:03 kernel: usb.c: USB disconnect on device 00:04.0-0 address 1 Jan 1 01:00:03 kernel: usb.c: USB disconnect on device 00:04.0-1 address 2 Jan 1 01:00:03 kernel: usb.c: USB bus 1 deregistered Jan 1 01:00:03 kernel: usb.c: deregistering driver usbdevfs Jan 1 01:00:03 kernel: usb.c: deregistering driver hub Jan 1 01:00:03 kernel: eth0: Broadcom BCM47xx 10/100 Mbps Ethernet Controller 3.90.37.0 Jan 1 01:00:03 kernel: eth1: Broadcom BCM47xx 10/100 Mbps Ethernet Controller 3.90.37.0 Jan 1 01:00:03 kernel: PCI: Enabling device 01:02.0 (0004 -> 0006) Jan 1 01:00:03 kernel: eth2: Broadcom BCM4320 802.11 Wireless Controller 3.90.37.0 Jan 1 01:00:03 kernel: device eth0 entered promiscuous mode Jan 1 01:00:03 kernel: device eth2 entered promiscuous mode Jan 1 01:00:03 kernel: br0: port 2(eth2) entering listening state Jan 1 01:00:03 kernel: br0: port 1(eth0) entering listening state Jan 1 01:00:03 kernel: br0: port 2(eth2) entering learning state Jan 1 01:00:03 kernel: br0: port 1(eth0) entering learning state Jan 1 01:00:03 kernel: br0: port 2(eth2) entering forwarding state Jan 1 01:00:03 kernel: br0: topology change detected, propagating Jan 1 01:00:03 kernel: br0: port 1(eth0) entering forwarding state Jan 1 01:00:03 kernel: br0: topology change detected, propagating Jan 1 01:00:03 dnsmasq[82]: started, version 2.22 cachesize 150 Jan 1 01:00:03 dnsmasq[82]: DHCP, IP range 192.168.1.4 -- 192.168.1.250, lease time 24h Jan 1 01:00:03 dnsmasq[82]: DHCP, /tmp/dnsmasq.log will be written every 8h Jan 1 01:00:03 dnsmasq[82]: read /etc/hosts - 5 addresses Jan 1 01:00:03 dnsmasq[82]: read /etc/ethers - 3 addresses Jan 1 01:00:03 dnsmasq[82]: reading /tmp/resolv.conf Jan 1 01:00:03 kernel: usb.c: registered new driver usbdevfs Jan 1 01:00:03 kernel: usb.c: registered new driver hub Jan 1 01:00:03 kernel: usb-ohci.c: USB OHCI at membase 0xb8004000, IRQ 2 Jan 1 01:00:03 kernel: usb-ohci.c: usb-00:04.0, PCI device 14e4:4715 Jan 1 01:00:03 kernel: usb.c: new USB bus registered, assigned bus number 1 Jan 1 01:00:03 kernel: hub.c: USB hub found Jan 1 01:00:03 kernel: hub.c: 2 ports detected Jan 1 01:00:04 kernel: hub.c: new USB device 00:04.0-1, assigned address 2 Jan 1 01:00:04 kernel: usb.c: USB device 2 (vend/prod 0x930/0x653d) is not claimed by any active driver. Jan 1 01:00:04 kernel: lp0: using parport0 (polling). Jan 1 01:00:04 kernel: usb.c: registered new driver usblp Jan 1 01:00:04 kernel: printer.c: v0.13: USB Printer Device Class driver Jan 1 01:00:04 kernel: SCSI subsystem driver Revision: 1.00 Jan 1 01:00:04 kernel: Initializing USB Mass Storage driver... Jan 1 01:00:04 kernel: usb.c: registered new driver usb-storage Jan 1 01:00:04 kernel: scsi0 : SCSI emulation for USB Mass Storage devices Jan 1 01:00:04 kernel: Vendor: 0s €Hû € Model: ôÿÿÿ Ú*€´– Àõÿÿÿ Rev: ™€ Jan 1 01:00:04 kernel: Type: Direct-Access ANSI SCSI revision: 02 Jan 1 01:00:04 kernel: Attached scsi removable disk sda at scsi0, channel 0, id 0, lun 0 Jan 1 01:00:04 kernel: SCSI device sda: 1971200 512-byte hdwr sectors (1009 MB) Jan 1 01:00:05 kernel: sda: Write Protect is off Jan 1 01:00:05 kernel: Partition check: Jan 1 01:00:05 kernel: p1 Jan 1 01:00:05 kernel: USB Mass Storage support registered. Jan 1 01:00:05 udhcpc[100]: udhcpc (v0.9.9-pre) started Jan 1 01:00:06 dnsmasq[82]: read /etc/hosts - 5 addresses Jan 1 01:00:06 dnsmasq[82]: read /etc/ethers - 3 addresses Jan 1 01:00:06 dnsmasq[82]: reading /tmp/resolv.conf

Some idea?

PS: Things has become worst since I can not boot from it any loger but everything seems Ok with the pen drive.

I have been doing further tests with my pen drive. Latest is that I have installed Linux Minime on it, boot it up from my PC and it works. I have done it a couple of times without any sort of problem.

I have been testing and thinking of this problem without any successful result yet. It could be that the USB needs a longer time to get stabilized after a reboot. Do you know if there some parameter in the nvram to produce an extra delay before the router searches for the bootloader?

I have boot_wait=on but it does not help.

After upgrading to 1.9.2.7-7f I got the USB pendrive to boot again. It is the same pendrive as before.

Check the USBsticks integrity first

With the WL500 open to Linux, we tend to try to install all the services of a regular Linux desktop. USBsticks are not really equivalent of a rotating hard disk or DRAM, they are solidstate and prefer to function with minimal write cycles and mostly read cycles.

If you installed a RAMdisk on the USB, it adds a big wear factor. And even having the Linux system in a RW mode, made increase wear. Nobody knows for sure, but at some point the use of the WL500 with Linux requires some careful thought about when and how to use the solidstate memory for RW mode.

If you indeed wore out your USBstick, tell us, and provide specifics, as this is important information.:o

i think i have the same problem

http://wl500g.info/showthread.php?t=15694

Well, since the firmware upgrade I have not had any problem with the USBstick. Last January I upgraded the system from the old 1GB USBstick to a new 8GB one. I'm still using the old one as an ordinary portable USBstick. The new one is working perfect... so far. I'm aware of the ware out issue but I rather pay the price than having a hard drive spinning 24/7/365. For those that like hacking, I have tried a Deskstar drive which I configured to spin down after 20minutes (Hitachi tools). It is possible to get the drive to spin down if the syslog is saved on the ramfs (/tmp) instead. If you want to keep your syslog after warm reboots you can always use the pre-shutdown script and copy it to the USBstick/hard drive.

Regarding “bigphantom” post, I never lost information on the USBstick; I just got unstable boots –sometimes it booted from the USBstick sometimes not. But of course when the router was booting from the internal memory then none of the configuration found on the USBstick was available; that is, the router was not configured as on the configuration files and/or scripts I had on the USBstick.