Hi Stefan, Thanks for your comments. This weekend I was given time by my wife to investigate the problem I had. I finally think I found the solution to my problem.
Here is the config a) Two hard disks, 1 disk PATA, 1 disk SATA and 1 CD/DVD Read/write PATA. That means that between the PATA devices, one becomes master, the other the slave. The USB backup device I had had a IDE (PATA) drive, and it needs to use drive select to determine if it is to emulate a master or a slave. The interface tries one or the other and fails, or collided with the state of the CD/DVD, ergo problems. Sometimes the drive would lock up, other times it would be slow. When I unplugged the IDE hard disk, this freed up a line for master/slave, and my external hard disk just worked normally. When I repowered the device, but removed the CD/DVD ROM from the bios, the USB drive worked OK. Based on my experimentation, I believe that the USB driver builds a bridge to the IDE device driver interface. Any IDE command gets transfered via the USB port to the external drive. I am now looking to replace the external USB box with a SATA hard drive. The USB backup box should have either eSATA, firewire or USB interface. Tuesday, Fedora 14 is out. I will wait a few days before downloading the DVD. As I started with Fedora, and even though I have UBUNTU on one machine in the office, I still feel more comfortable with the former. I want to see if my problem is Fedora or UBUNTU related. By the way, on this computer where I was having the problems, I have quadruple boot, XP, Windows 7, Fedora and UBUNTU. All for some sofware development or QA activities. ------------------ Regards Leslie Mr. Leslie Satenstein mailto:[email protected] mailto [email protected] / [email protected] www.itbms.biz --- On Sat, 10/30/10, Stefan Monnier <[email protected]> wrote: From: Stefan Monnier <[email protected]> Subject: Re: [MLUG] BTFRS for USB Hard Drive To: "Montreal Linux Users Group" <[email protected]> Date: Saturday, October 30, 2010, 9:04 PM > In a few earlier postings I indicated the problems I had with EXT4 and > a hard drive that was about 75% full (320gig at 75%). Performance was > such that simple uploads or downloads crawled to a 10k / second pace, > or even stalled for long periods of time. That sounds like an ext4-specific problem. Most likely any other filesystem would not show such a problem (tho all filesystems do slow down to some extent as the free-space goes down, but 75% is generally considered as "almost empty" in this respect). And even ext4 would probably not show this problem if you could figure out what was the cause of the problem or if you upgraded to a slightly more recent version of ext4 (after all, it's still a work-in-progress). > Ergo, if someone inadvertantly pulls out the drive's plug from the USB > physical port, we are almost guaranteed to not have drive corruption. > There may be some data loss, but no drive corruption. That to me is > most important. "drive corruption" means that the drive itself has some corruption, which is unrelated to the filesystem (or partitioning) you use on the drive. So you probably mean "filesystem corruption" instead. Note that most/all recent filesystems provide such a "guarantee" (modulo bugs): ext3, jfs, xfs, you name it. > I am not certain about storage efficiencies. Would 10gigs of data > with EXT4 use less diskspace than 10 gigs of data stored in btfrs > format. (What's a gig when terrabyte drives sell for around $60.00. There are always some slight differences in this respect, but they should usually be lost in the noise (a few percents at most). Stefan _______________________________________________ mlug mailing list [email protected] https://listes.koumbit.net/cgi-bin/mailman/listinfo/mlug-listserv.mlug.ca
_______________________________________________ mlug mailing list [email protected] https://listes.koumbit.net/cgi-bin/mailman/listinfo/mlug-listserv.mlug.ca
