UltimateDefrag4

I own two licenses to UltimateDefrag 4 and I am having trouble on both machines.

I wish to defrag a large (2TB, 32% full) internal non-system HDD, but cannot progress past the 'Analyzing' stage: at 88% completion, the program terminates abruptly and the window closes. This occurs in both machines (with very similar hardware and software configurations) and regardless of the various options and settings I choose. A Crash.dmp file is produced in each case but I don't know how to read it or what to do with it.

Running Windows chkdsk (all 5 phases) made no difference. I also ran PerfectDisk. This resulted in UltimateDefrag Analyzing taking less time to reach 88% but once it did, it terminated as before.
I have a second large (2TB, 36% full) internal non-system HDD installed, but here UtimateDefrag works just fine.

Customer support at DiskTrix does not respond.
Any help would be greatly appreciated. (Windows-XP-Pro-SP3-32bit)

Norman

 

have had similar issues with ud and usb drives. and same for me no one from disktrix responds anymore. i find ud has issues and some that were never resolved and they seem to have not ever made an attempt to do so. anyway..... i do not currently have it installed but can you tell which file it starts to read right before this happens? i had a similar issue and i ended up having to remove the file ud did not like and then it analyzed and defragged fine. imo there is a file its hitting and getting hung up on. disktrix support was horrid and they imo could care less if you paid good money to support them. personally i would switch to perfectdisk (or others puran defrag is great but be careful during the install for the extra junk) and call it a day.

 

I believe the company to be, for all intents and purposes, disbanded from a consumer point of view. No support, no revisions or updates. I feel the company had put together a product, ran into internal problems, disbanded, and left the site up as a source of income. The website seems to be in the "care" of someone that is collecting the funds from an occasional purchase. And support is non-existent. Totally.

You're just seeing a store front that's been left standing.

As much as like the idea of custom file placement and all that, I CANNOT, and will not recommend it.

Regarding the technical problems, I feel that UD is fully functional and works correctly within limits. Limits which are not clearly defined. And everything has to be "in order" for a defrag to complete itself.

Getting BootTime defrag (for metafiles) to work is a tricky Dick too. For one thing, when you exit UltimateDefrag after having selected BootTime defrag, you need to wait a minute for the startup process to be reconfigured. And the machine's disk configuration has to be the same at defrag time as when you're settings the BootTime options. This means no connecting or unplugging USB disks during the reboot. And the disk(s) that are to be operated on must be spinning and ready to go. Some Eco disks and USB disks with overly aggressive power savings may fail here.

Also, if there are any unreadable sectors on your disk, this will crash, or lock-up the program. This is not necessarily bad (that UD would fail here) because there is a disk problem. Investigate the problem file thoroughly. And if any of the disk's metadata is out of order, this can cause RunTime or BootTime operations to also fail.

DiskTrix could polish all this up and present clear error messages instead of simply crapping out or aborting.

I also believe that UD is only suitable for XP, and questionable with Vista/7. Forget 8.x completely.

 

The idea of a bad file seems an excellent place to look. But how can one tell? The System logs are useless, _Defrag.log (evoked via Ctrl-Shift-F12) appears and disappears beyond my control, and no files are actually moved during the Analyze phase (so that there are no flashing 'moving to' or 'moving from' squares). Is there another way to find which file UD is reading when it terminates?

I do have Perfectdisk (version 10, but I would gladly upgrade to 13 if I thought it would help), but what I wanted UD for was to reposition my folders on my harddisk. The only other program I could find that would do this, is Auslogics Disk Defrag Pro but it is not as easy to set up and not nearly as flexible as UD and is very very slow. I don't even mind having to use two defrag programs, one to reposition the folders, the other to defrag the files. Any suggestions would be most welcome.

Norman

 

Pity, because when it does work, it works well.

I ran chkdsk and Auslogics Disk Defrag Pro's checkdisk and they found no problems.

I only run XP in any case, so for me that is not a limitation.

Norman

 

Norman, both MHDD and HDATA2 are decent surface scanners and should be able to detect a flaw in the file itself... and both "claim" to be able to do BAD SECTOR repair.

Probably worth a try... after a FULL image, of course

 

No such thing as a bad sector repair in which the data is saved. The sector can be repaired, but the data it held is gone! El'finito! The file will be corrupt.

First thing I'd do is scan disk sector-by-sector to see if there are any bad ones.

 

I typically use HD Tune Pro, just anything to read each and every sector on the disk and report ones that are bad.

 

Thank you both. Since UD terminates abruptly at around the same place in both machines, and both contain much the same data, I am assuming that the corruption lies in the file not the disk.

Is that reasonable? If so, would MHDD find the errant file? I have copies so if the file gets zapped in the process, that's okay.

Would HD Tune do what I need or do I require the Pro version?

Norman

 

HD Tune comes with a 15 day trial period. Both the standard and pro version will non-destructively read each sector. There are other freeware tools that can perform this function too, as mentioned. I've just been jamming with HD Tune for so long it's like old leather.

All I'm wanting to see is that each and every sector can be read error-free. And since we don't know anything about UD is doing internally we can only eliminate things step-by-step. I can tell you that if it crashes during the analyze phase that the problem is most likely with how a file is recorded, something internal to NTFS. Because, during the analyze phase the complete file is not read. Reading each file completely could take hours.

Describe your hardware setup, cpu, memory, drive make and model, interface, any other software packages running in the background? How many files total on each disk?

By moving files in small groups to another disk for safekeeping, you can eventually find the troublemaker. If you get down to zero files and it's still doing it, then you've got a disk structure problem. And if after formatting it, it still does it then you've got capacity or hardware incompatibility/failure.

Tedious? Yeh. Non-elegant, yep. Anyone else have more specific suggestions? Let's hear'em.

 

Thank you Keatah for your willingness to help and the detailed roadmap you provided.

I ran HD Tune (free) Error Scan (not Quick Scan); it took 4 hours and found no damaged blocks.
I'm now beginning the process of moving files to another disk and trying to isolate the troublemaker.

Hardware setup: system disk (2TB) with 4 partitions (C:, D:, E:, F: , 2 storage disks G: and H:
Problem disk is H:. It is a 4TB WD SATA II drive formatted for XP as a single 2TB partition; so is G:
CPU: x86 Family 6 Model 30 Stepping 5 GenuineIntel ~2793 Mhz
Total Physical Memory 8,192.00 MB
Drive: WDC WD40EFRX-68WT0N0
Interface: SATA II
Other software in background: McAfee

Bingo! You did it. I made room on G:, transfered most of H: to G:, got rid of the rest.
Now UD terminates abruptly on G: as well.
Total Files G: 1,548,501 (89%), terminates at 90%
Total Files H: 1,541,713 (83%), terminates at 84%

In Jerusalem we work a five-day week too, but Sundays to Thursdays, and since we are at GMT+3, it's getting near quitting time. I do have remote access from home, but this is the kind of work I feel safer doing from the local console, so I probably won't finish until next week (Sunday). Until then, thank you again.

Norman

 

I don't know if UD is failing because of a capacity limit, perhaps some condition related to file quantity or total combined size is causing it to fail. Maybe there is one file that is too large or contains an illegal character in its filename? Maybe McAfee is holding up something? Could one of the NTFS metafiles have grown in size beyond what UD can handle?

Lots of suspicions, but without understanding UD's internal workings and limitations I'm still guessing.

 

Things are a lot clearer now, thanks to you. It is not a particular file that is causing the problem nor the amount of data on the disk as such but rather the number of files it contains.

I doubt that we can extend UD's limit. But there are a few possible workarounds. I could create two partitions on the offending disk or I could zip some of its directories. (I am assuming that UD counts a zip file as one.)

Let me avail myself of your kindness (and superior knowledge) to present another problem—actually, the reason for trying to use UD in the first place. When I do a full backup of my system disk using Macrium Reflect (about 65GB with standard compression) and store it on G:, it takes about 20min; on H: it takes about 40min. Since G: and H: are identical physically and contain nearly the same amount of data, I thought it might be because of the location of the backup directory on the storage disk. Hence UD.

During preparation for the testing you proposed, I copied a backup file from H: to G:. At first, for about half the file, the copy speed was around 120MB/sec. Then, suddenly, it dropped to below 10MB/sec and remained there for the rest of the copy. Nothing else was running.

I ran a Benchmark with HD Tune (free). With G: it was reasonably uniform at about 147MB/sec, dropping off somewhat at 65% (minimum 112, maximum 166, average 139). With H: the trace oscillates wildly, from a minimum of 12.5MB/sec to a maximum of 155.4 (average 116).

Can H: be repaired? Are there any other tests I should do? All the stuff on H: is now backed-up elsewhere, so the testing can even be destructive. How about just a format? I am prepared to buy a new disk if there are no alternatives.

What do you think?


Norman