Am 26.05.2011 22:41, schrieb Benoît Minisini: >> Sometimes bad power cable connection does things like that, but maybe >> Benoit has already check those. >> >> Jussi >> > > I come back, with a new firmware. > > I have two SATA hard disks. The first one is the original one, and I had no > freeze with it. The second one, the WD Velociraptor, was added after. When I > got the freezes, I swapped the two cables, but it changed nothing. > > Concretely, the drive stops between 1 and 2 mn each hour, making the system > unusable until it wakes up. Only the mouse and processes doing no disk access > work. Sometimes I have no freeze at all during a long time. > > For the SMART utility, there is no problem at all on the disk. > > I noticed that running Firefox help having the freezes. So I suspected the > kernel part of the nouveau driver. But I use the NVIDIA blob since a few days, > and I got the freezes too. They are just shorter (?) > > Then I found on the web someone that created a bootable CD with Windows 98 > that updates the firwares of my model (WD1500HLFS). On the web page, some > comments of a guy that describes the same problem as mine telling that the > update fixed it! So... > > Now, let's see if I will continue having freezes... >
Arrgh - if I had seen that earlier, I might have warned you it could be futile. Currently, there is a discussion between the kernel guys going on about this, it is due to a kernel module which was changed to make it fit for a special controller but it had this "side effect" on other controllers. I stumbled across this because I had the same problem on my laptop. In the Linux forum I use, someone pointed me to it as a possible solution. I used iotop to find the source of the writing activity, but to no avail. But tail -f /var/log/messages shows errors on SATA each time the system comes back. First I thought of the hardware, but that guy pointed me to this discussion: One Posting says that "device reported invalid CHS sector 0" could come from a kernel bug: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/397096 The line: "device reported invalid CHS sector 0", seems to be an "off by one error" reported in http://patchwork.kernel.org/patch/41773/ and fixed upstream, and not seen often "The bug isn't very visible because ata_tf_read_block() is used only when generating sense data for a failed RW command and CHS addressing isn't used too often these days.", Tejun Heo says. Then he said the problem might be in /usr/src/linux-2.6.37.6-0.5/drivers/ata/ahci.c which doesn't help me too much ;-) And he gives this URL http://help.lockergnome.com/linux/AHCI-driver-problem-SB700-SB800-Acer-Ferrari--ftopict537147.html Now, what do you think? Could this be your problem, too? If so, I guess we just have to wait for the next kernel update from the repos? Regards Rolf ------------------------------------------------------------------------------ vRanger cuts backup time in half-while increasing security. With the market-leading solution for virtual backup and recovery, you get blazing-fast, flexible, and affordable data protection. Download your free trial now. http://p.sf.net/sfu/quest-d2dcopy1 _______________________________________________ Gambas-user mailing list Gambas-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/gambas-user
