Jump to content

cosmotopper

Members
  • Posts

    6
  • Joined

  • Last visited

Reputation

0 Neutral
  1. Thanks to Piriform for a great lineup of elegant and excellent tools. At the risk of sounding paranoid (and based on 30 years experience dealing with the most pernicious monopoly in the history of US business practice), the problem with these two files isn't that they are, it's where they are. I have been trying to defrag my NTFS volume so I could shrink it down from 640GB to 40GB, so I could make room for installing a dual-boot Solaris10 on this machine. Unfortunately, whenever I try to use the Vista supplied 'Shrink' functions, they inform me that the minimum size I can shrink it to is around 300GB. The folks at Microsoft don't really want anybody elses operating system on my PC. That's because they don't really consider my PC to be 'my PC', it's really their PC, and they're just letting me use it. I don't have any problem with them wanting to put a hibernate file, or a system restore file, or a big paging file on the volume. But why design a file system so the data clusters can't be organized any way an appropriately authorized admin wants them organized? This is no coincidence. These log files showed up (in the middle of a nearly empty drive) only after I turned off the drive 'journaling' function, after I had removed the hibernate file, etc. etc. Yes, it's possible to turn that stuff off, but only after wading through the byzantine and self-obfuscated so-called 'knowledge base' provided by the mega-geeks up north of us here in Portland OR. Hey Steve (Balmer), All I want to do is shrink the NTFS volume down a bit, I'm not kicking Vista totally off the machine. Why make it so damn difficult? Thankfully, 2-minutes with a Linux live-CD and I can fix it anyway, but why should I have to do that? Thanks again for the great work you folks at Piriform do. I donated, I hope everyone else who uses these great utils does also.
  2. (IMHO) Here are the crucial features defraggler is missing (I think): 1. We should be able to not only defragment, but compact data files at arbitrary boundaries (meaning, group them all in one place). The default would be to compact all movable data at the bottom of the disk, leaving as much free space as possible, as one would need to do in order to resize the partition from larger to smaller. 2. Ideally, given the ability to compact files, we would like to be able to divide the total disk space into two zones, a static zone at the base, with all other folders and files compacted either at the end of the disk, or at some arbitrary offset from the top of the static zone. It should also be possible to designate a pair of files as the bookends for each zone (one at the beginning, and the other at the end). The rationale here is that this schema would enable the creation of a tripwire process at the driver level, which would detect any attempt to write into the base zone. If, for example, all of the static operating system files (files which should never change), and optionally newly installed application files (which should not normally change either) were situated in this monitored base zone, it would be quite easy to detect attempts by malicious programs to infect files which should not be changed, by monitoring write operations below a certain offset. It would also enable defraggler to optimize subsequent operations simply by verifying that files in a given zone were still in the proper place, given a list of the original static files. Of course this feature would have to work around the inherent defects in the NTFS file system, but it would provide a powerful tool for managing the disk and detecting malicious activity. (It would also be nice if Piriform could write the monitoring program as well.) I realize this bleeds over into the 'anti-virus' application area. However, I consider most anti-virus software to be a virus in and of itself, because it intrudes on the operation of the system and often interferes with legitimate applications. Therefore a very simple monitoring program which would detect any attempt to compromise the core operating system files, used in concert with something like Windows Defender, would eliminate the need for these overly complex and intrusive virus-detection systems, at least for some users. As far as I know, the sources for your programs are not publicly available, but I'd be willing to work on these extensions under an NDA, pro-bono if you have an interest, but not the resources to prioritize this yourselves. Piriform is doing a very good job of maintaining it's focus on reliability and providing a simple, consistent GUI. Your products are simply outstanding, thank you! (I'm pretty sure I made a donation quite some time ago probably under CCleaner, but if not, please write back and let me know.)
  3. When I originally installed Defraggler, I was very pleased (so much so, I prematurely sent a donation). It appeared to be doing a great job. Unfortunately, having processed a few more devices, I'm simply mystified as to the inconsistent and seemingly ill-considered protocols. To highlight this, I'll use one test case which leaves no room for doubt: I have a 160GB USB drive formatted NTFS. There is 66GB in use, slightly more than one third of the total. Probably 20 of the files are 1GB or more. Given this scenario, it should be possible to defrag every single file. I could defrag the drive simply by copying all 66GB onto another drive, erasing it, and then copying the files back one by one (caveat, some copy programs might needlessly interfere in that process, leaving some fragments). However, after running a basic pass, I ended up with about 50 fragmented files and a little over 1000 fragments. So, I set the options to force files 250MB or larger to the end of the disk. That seemed to reduce the number of fragments a little bit, but at the end, the graphic display showed what appeared to be the end of the disk with a solid block of red clusters(?). The number of fragmented files had been reduced by about 10. There was plenty of free space (white?) on the display, so I'm wondering, why didn't it defrag the remaining files? Next, I reduced the limit on the 'move to end of disk' to 25MB. Since I had already moved the 250MB+ files to the end of the disk, I would have assumed that Defraggler would leave them where they were, and simply add new ones where it left off packing the first group. Instead, it appeared to refragment the first set of files in order to move the smaller files into their place. This sounds absurd to me. UPDATE: After some investigation, I find that the NTFS file system design is (predictably, since it's a product of Microsoft) seriously flawed. Clearly, using a large USB external device (or any large drive for that matter) creates major conflicts if one were to attempt to store (lets say for example): one branch containing lots of files, many small documents, images etc., along with some number of large multimedia files, DVD ISOs, or ZIP archives. In the above test case, I find that my 'MFT' has 18 fragments, which 'cannot be defragmented' (according to the Microsoft documentation). This is absurd, hence explaining the absurd results referenced above. Apologies to Piriform for the inuendo that they might be responsible. FYI, I plan to navigate around this problem by re-formatting all my NTFS external media with at least one virtual drive (either ISODisk or TrueCrypt at the moment) which should have the effect of reserving a large contiguous 'super-cluster' which can be used either for big multimedia/zip files, or lots of little files, while not overloading the brilliantly conceived Master File Table (which can't be resized, moved or defragmented). (I can only say thank God for the European Union's legal team for going after this criminal organization (stock symbol:MSFT) which has fed off the Personal Computer industry almost since it's inception.)
  4. I recently got nailed with a virus which infected a bunch of system files, and possibly some of my own. I wish I could have gone back and found where it had written stuff and which files it had contaminated. It would be useful to be able to put the static Windows files and folders at the top of the disk, and do an MD5 on all of them while you're at it. You could carve out space for the pagefile and hiberfile and use them as a buffer between that which is static, and the rest of the disk. Once optimized, you could very likely provide an optional utility which would detect any attempt to write new data into the static zone. You could also provide an option for users to configure the 'static files and folder' template to include new stuff (we) install later, and want to make sure it's left untouched. The MD5 option would enable Defraggler to periodically check the static area to see if anything was changed. This wouldn't really be all that hard to do: just provide Defraggler with a '-static_list:<file>' option, and have it go to work re-positioning things. Also, for general optimization features, add a folder in the DF install directory which could hold 'spacer files' placed at strategic locations, thereby enabling it to quickly re-optimize pre-defined disk regions by releasing spacers when doing so would avoid having to move a very large file. (Obviously there would have to be a user friendly 'clear spacers' option somewhere in the mix.) Windows system restore is one of the few things they've done which I've liked. However, it is still a very cumbersome system, and it has a primary defect in that it dumps restore points off the back end. Furthermore, in my case, the malware wiped out the restore data. If you implemented the above, Defraggler could provide the ability to snapshot the system periodically, and then store a perfect copy of the snapshot on an external media, which could be used to restore the system to perfect working order later in the event of a problem.
  5. I'm a longtime user of CCleaner (and a big fan of Piriform), who just started using the Fraggler. Not to toot my own horn, but I wrote the first commercial defrag utility for MS-DOS back in 1985, so I know a little about this particular app. Piriforms utility is outstanding. However... I just defragged an external USB 1TB Seagate drive with about 380,000 files, using roughly 300GB of the drive space. The progress percentage proceeded nicely, and finally got to 100%, but kept going for another half-hour or so. It had reported one remaining file with 2 fragments. Once it hit 100%, I had no way of knowing how far it had gotten, or what it was doing exactly. If there's just one more file to defrag, why not do it and stop? I had no way of knowing when it would finish. I also noticed that there were a handful of cluster tokens which were still sort of scattered, so I assumed that maybe it was trying to consolidate as well as defrag, which is fine, but it would be helpful to know. Then, because I was trying to fix an unrelated Vista F.UP while I waited for DF to finish, I accidentally hit the reset button... Upon rebooting, I read your docs and I guess I'll take your word for it that Microsoft's Defrag API is sound (although after nearly 30 years experience with thousands of defects in every OS they have ever produced, I'm still a little worried). I'd feel better if it were possible to do a filesystem integrity check of some kind (or maybe the Analysis step does that already). (?) This is a minor point, but I thought it worth mentioning. IMHO, Piriform is one of the most outstanding freeware providers I've seen. I donated, and anyone else who's using these elegant and finely crafted tools should too.
  6. I have a Toshiba Laptop running Windows XP SP2 and it appears that CCleaner (1.39.502) is erasing restore point(s). I noticed that they were all gone after attempting to recover from an unrelated install. Just to be sure, I set a new point, verified that it was ready/available, then ran CCleaner "Clean" option. I found no reference to the restore point files in the list, so I erased them. Immediately went back to system restore and found that the point I had set was gone. Next I checked my system restore points on a nearby Gateway 4016. Lots of points in place, but not sure when I ran my last CCleaner "Clean". Ran it again, cleaned it, and the restore points remained intact. I don't have Norton or shredder or any of the other stuff going on the Toshiba, and since this was an immediate test (set restore point, run CClean, restore point gone), it suggests there may be some issue. I'll keep observing this to see if I can narrow it down. CClean is an outstanding program, but if it has this vulnerability, I'm sure MrG will want to fix it.
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.