September 29, 200916 yr Author I hate not solving problems, so...... I took the ram out of my main PC and installed it in the server ( both are the exact same type ). It passed the memory test, with both sticks in. That at least answers the "is it compatible" question. I ran the memtest 3 cycles with no issue. I can run a more extensive test later, however with the other ram it always failed within 5 minutes. I am not re-running the burn-in on both drives, however...... I even with the "bad" ram, my first test passed ( as posted earlier ). I may have to run the test a couple of cycles before I am confident the ram is ok with it. The other ram... well..... I using it on the PC I am typing with right now. I'll force the RMA issue with OCZ once I add to the trouble ticket that the other pair of ram works without issue ( after testing it over night tomorrow ). Hopefully by the weekend I can start working on the software/data. Speaking of........I should follow up with Tom tomorrow about my license I bought off another user. Stress level is down, now that the other ram is working at least. Time to go to bed before something goes wrong Thanks everyone for the continual feedback. And Joe... glad to hear my failure has sparked a new idea for you. I promise to provide you with many more in the days to come! hehehe
September 29, 200916 yr I hate not solving problems, so...... I took the ram out of my main PC and installed it in the server ( both are the exact same type ). It passed the memory test, with both sticks in. That at least answers the "is it compatible" question. I ran the memtest 3 cycles with no issue. I can run a more extensive test later, however with the other ram it always failed within 5 minutes. That sounds like some good news. I am not re-running the burn-in on both drives, however...... I even with the "bad" ram, my first test passed ( as posted earlier ). I may have to run the test a couple of cycles before I am confident the ram is ok with it. Problem is, we have no way of knowing if the drive really is zeroed. Did the RAM corrupt the value you "wrote" to the drive, so it actually has bit not zeroed out? or was it written fine, and the bad RAM caused the read operation to mis-read every once in a while? Let me give a tiny bit of thought to it and I think if you post a copy of the syslog I can give you a one line command to verify a spot or two that were mentioned in the "bad" results of the one disk. The other ram... well..... I using it on the PC I am typing with right now. I'll force the RMA issue with OCZ once I add to the trouble ticket that the other pair of ram works without issue ( after testing it over night tomorrow ). Hopefully by the weekend I can start working on the software/data. Speaking of........I should follow up with Tom tomorrow about my license I bought off another user. Stress level is down, now that the other ram is working at least. Time to go to bed before something goes wrong Thanks everyone for the continual feedback. And Joe... glad to hear my failure has sparked a new idea for you. I promise to provide you with many more in the days to come! hehehe For now, you can blame any crashes on Microsoft , and not the RAM strips.
September 29, 200916 yr Author OOpps... When I said I am "NOT" reruning the burn-in, that "T" should have been a "W", meaning "NOW". Funny how one letter can completely change the meaning of a sentence. I'm running the process now and it should be completed around 9pm tonight. I'm confident it will come back clean (positive thinking now). Once it does, I will run the memory test over night , on my old RAM. The new RAM.... I will fight with it later. As you said, I can blame Windows if my main machine crashes.
October 12, 200916 yr Author I've just put in 3 used drives into my server. When I look at the drives in UnMenu I see this: Drive Partitions - Not In Protected Array Device Model/Serial Mounted File System Temp Size Used %Used Free /dev/sdc scsi-SATA_ST3750330AS_5QK00GT5 30°C 750.1G /dev/sdc1 scsi-SATA_ST3750330AS_5QK00GT5 ntfs 30°C /dev/sdd scsi-SATA_ST31000528AS_9VP0H3TD 26°C 1000.2G /dev/sdd1 scsi-SATA_ST31000528AS_9VP0H3TD ntfs 26°C /dev/hda ST320011A_3HT0GTF1 30°C 20.0G /dev/hda1 ST320011A_3HT0GTF1 ntfs 30°C Why does it see each of the drives as both sdx and sdx1 ? I just deleted the files on the drives on my windows machines before I put them in the server. Do I need to "prep" these drives before I start the preclear process? ( all the other drives I have used, until now, were new ). If I don't need to do any other prep, then I would only run 3 preclear commands as: preclear_disk.sh /dev/sdc preclear_disk.sh /dev/sdd preclear_disk.sh /dev/hda ( I was thinking of using hda as my cache drive... that's why I am putting in such a small drive )
October 12, 200916 yr I've just put in 3 used drives into my server. When I look at the drives in UnMenu I see this: Drive Partitions - Not In Protected Array Device Model/Serial Mounted File System Temp Size Used %Used Free /dev/sdc scsi-SATA_ST3750330AS_5QK00GT5 30°C 750.1G /dev/sdc1 scsi-SATA_ST3750330AS_5QK00GT5 ntfs 30°C /dev/sdd scsi-SATA_ST31000528AS_9VP0H3TD 26°C 1000.2G /dev/sdd1 scsi-SATA_ST31000528AS_9VP0H3TD ntfs 26°C /dev/hda ST320011A_3HT0GTF1 30°C 20.0G /dev/hda1 ST320011A_3HT0GTF1 ntfs 30°C Why does it see each of the drives as both sdx and sdx1 ? The entire drive, including the partition table and boot record is accessible under linux as /dev/sdX (or /dev/hdX if PATA) The individual partitions on a drive have the base name of the partition with a trailing number, so /dev/sdx1 is the first partition on drive /dev/sdx, /dev/sdx2 would be the second partition, etc. unRAID uses only a single partition, as does windows in most cases. All those with the trailing "1" show as having an ntfs file system. Even though you removed the files, the file-system is still there. The preclear script needs the name of the entire drive, not a partition on it, so ignore the entries it is showing in unMENU with the numeric suffix, those are the file-system partitions. Upon being cleared, the preclear script will erase the existing partitioning and set up a unRAID compliant partition as part of the pre-clear signature. I just deleted the files on the drives on my windows machines before I put them in the server. Do I need to "prep" these drives before I start the preclear process? ( all the other drives I have used, until now, were new ). You did not even need to delete the files... Trust me, the pre-clear does a LOT more to delete your files than windows ever does. (All windows does on a delete is to delete the directory entries that point to the files, the disk blocks holding the data from the files are still there. That is how most of the un-delete utilities work, by putting back the directory pointer. If I don't need to do any other prep, then I would only run 3 preclear commands as: preclear_disk.sh /dev/sdc preclear_disk.sh /dev/sdd preclear_disk.sh /dev/hda ( I was thinking of using hda as my cache drive... that's why I am putting in such a small drive ) Yes, that is it exactly. Run each command in turn, or open up three telnet sessions and run one in each. Joe L.
Archived
This topic is now archived and is closed to further replies.