herbalist

I haven't had the chance to verify it, but I think I've found one of the items draining the RAM and/or resources. I'd forgotten that I had Kerio firewall set to log outbound Tor traffic. Needless to say, the log had become quite large. Deleting the log gave me back 25MB of RAM and available GDI resources went up 8%. The odd part here is that the decrease in RAM usage didn't show up in any of the processes listed in MemLoad but does show in in the free physical memory.

Nice selection of tools. Thanks. Mijzelf According to that article, I'm set much higher than I should be at 512 but have not seen any instability. Lowering "TcpTimedWaitDelay" to 30 might be one reason. The errors have never returned. I have to believe the unofficial upgrades, primarily Revolutions Pack made that possible. I'm also wondering how much effect selecting "network server" in file system properties has here. Never took the time to check what it changes. This instance of Tor has run for 17 days now, still seems fine. Total uptime has reached 3 weeks. Both the available RAM and GDI resources have gone back up some. I still wonder why the error messages both ended with "Not enough ram?" when MemLoad showed that I had RAM available. It also has me asking if stronger hardware would help here? Looking at the other Dell with 2GB RAM and a 2.6GHZ P4 sitting next to this unit, wondering if the other half would notice if I swapped hardware with her.

dencorso That does look useful. Thanks.

From what I've seen there, all of the non-server versions of Windows don't hold up well. 98 is especially bad "out of the box", but with the upgrades here it's been a very pleasant surprise. Yes, Linux would be a better choice but I don't have a free PC of any real power available to put it on. For me, Linux is like starting over, something I really don't have the time or ambition to do. In this regard, my goals conflict. I might not be able to get 98 to run 6 months continuously, but everything I'm seeing says I can get one month of reliable service. I don't see any Windows units, save a couple of servers, functioning as exit relays listed on Torstatus doing any better. In that respect, one goal has been achieved, having 98 perform at least as well as the new versions of Windows. My traffic load is quite variable, ranging from 100MB to 2GB per day. At times there's nearly 200 established connections. I've restricted the bandwidth to 24 KB/s average and 32 KB/s maximum in order to leave enough bandwidth for the phone to work. Even at this setting, when the traffic peaks, the phone conversation cuts out. At a higher setting, it fails to ring or dial out at times. Ideally I should upgrade the DSL service but that doubles the price and doesn't double the upload speed. Not much of a selection where I live. Netcat does look interesting. I wasn't aware of such a tool for Windows. I don't leave Vidalia running as such. I was starting it when I needed to shut Tor down cleanly. Can Netcat send a password to Tor? As for the resource leaks, I'm not certain that Tor is what's leaking. That's a large part of what I'm trying to determine. Anyone know of a resource monitor for 98 that shows the load of each process?

KernelEx was installed long before I added Tor and was disabled for the Tor executable before I ever launched it. The only purpose I have for Vidalia is for the timed shutdown of Tor, last used when I updated Tor almost 2 weeks ago. Tor definitely works fine on 98 with a few tweaks, well enough that the browser on my other PC becomes extremely slow and the phone (uses a separate modem) gets erratic. Those tweaks were done some time ago and are listed in earlier posts. The question I'm trying to deal with is how long can 98 serve as an exit node and remain stable. So far, it's holding up, but GDI resources are slowly dropping, down to 49% at the moment. Memory usage is slowly going up. Memload shows Kernel32.dll is now using 9.9MB, far higher than normal. Free physical memory is down to 145.5MB. It's clear that there is a limit to how long 98 will remain stable as an exit node, but the tools I have are not pointing to what is specifically using up the memory and resources. The only "problem" I've experienced is those log entries. There was no visible problems at the time or since. Vidalia wasn't in use. I wasn't browsing through Tor that day either. I'm looking for better information than is contained in the pages I linked to.

A quick update. Real life issues have kept me from giving this and other projects/experiments the time they deserve. For the most part, I've kept Tor updated and left it running as an exit node. On January 7, I updated Tor-Vidalia to the current version, 0.2.2.35-0.2.15-1. I'm no longer using Vidalia to start or monitor Tor, choosing to start Tor directly. I assigned a set password and use Vidalia just for controlled shutdowns of Tor. I haven't been able to make Vidalia stable on 98 for an extended time. On my system, the Vidalia network map works inconsistently to start, then not at all after a couple days. NetMeter has proven to be an excellent replacement for Vidalia's traffic monitor. I've changed torrc to log events to file instead of relying on the Vidalia message log. ClockMon has solved the problem of the Windows time clock losing time, making it unnecessary to use Time Sync. The present version of Tor has been running for 12 and 1/2 days. The previous version ran for 5 before I shut it down and updated. All total, this 98 unit has been running as an exit node for almost 18 days now. TorStatus changed its status yesterday, listing it as a stable relay, after which the traffic volume tripled. Except for some minor graphics issues with the browser when it's maximized, there have been no observable problems, save being slow when the Tor traffic load is heavy. A couple entries in the Tor log caught my attention and are the main reason I'm adding this update. On the morning of January 18, before my exit node status was changed to stable, I saw the WSAENOBUFS error for the first time. The entry appears twice in the log: Jan 18 06:33:53.890 [warn] write() failed: WSAENOBUFS. Not enough ram? Jan 18 06:43:48.450 [warn] write() failed: WSAENOBUFS. Not enough ram? MemLoad showed I had 280MB of RAM available. The swap file hasn't been used. Both instances took place while I had a large download in process and had several tabs open in SeaMonkey. Neither Tor or Windows crashed and neither showed any problems afterwards. Unless I've misunderstood the Tor bugtrak pages regarding this error, I was under the impression that this was terminal to Tor, causing it to crash and required a reboot of Windows to fix, none of which happened. My understanding of this error and the underlying system is quite limited. Could someone more knowledgeable look at these Torproject pages and tell me if I've got it wrong or have the upgrades made it possible for 98 to recover from a normally terminal error and continue to function? https://trac.torproject.org/projects/tor/wiki/doc/WindowsBufferProblems https://trac.torproject.org/projects/tor/ticket/98#comment:74

Since this is a wish list, is anyone willing to restart development on this Open Source Fatsec Security Driver? Description: Emulate filesystem security on Windows 9x systems with FAT16/32 filesystems. Files, directories, and drives can be assigned read/write/execute permissions which are enforced by a kernel-mode driver. Configuration file is locked while the driver is loaded. This was discussed about 3 years ago here but there wasn't much interest at the time. Now that there's almost no AV support for 98, maybe it should get another look.

I have written an API that supports multi-core. At present it does require that Applications be written to use the API. I was looking at the idea of running Virtual PC (or a better option if Kex and Import patcher make that possible) on its own processor while the rest of the system runs on the first processor. Would this require a complete rewrite of VPC?

I'd like to add my thanks to Gelosea too. I installed the whole package, ignoring the IE6 recommended prompt. Originally I didn't plan on changing the appearance. Installed it mainly for improved resource handling. I doubt this unit could run a Tor exit node without RP9. I did like the black mesa theme but found that using any theme made some items unresponsive. The Tor traffic graph for instance stops moving. Without building an unstripped default system on the same hardware, I can't determine if it's a bug or the result of my stripping IE and more out of the system. I didn't report it as I figured Tihiy had enough to do without having to account for systems as altered as this one, especially when my changes may have caused the problem.

Doesn't look like it. It's hard to remember for certain but I don't recall the commands being predefined. I'm pretty sure the app was free or possibly a beta. The app worked on 98FE which is what I had at the time.

I've had a similar problem with toolbars. The dividers between several toolbars disappear if I lock the taskbar but don't reappear when I unlock it. For me, the work-around has been to unlock the taskbar, delete the toolbars, then put them back. When I add them back, the dividers and menu entries are there again. I always assumed it was due to my stripping out Internet Explorer, but since I don't lock the taskbar, it hasn't been an issue.

Back around 2003-2005, I was trying out several text to speech, speech to text, and voice control programs. There was one in particular, a voice control program that required you to train it to your voice. It was a small program, maybe a couple megabytes at most. If I remember, it added a control panel applet. It had an odd name, similar to IM2 or A2. In many ways, it reminds me of the Dragon speech software that's advertised now. It did most all those things. At the time, I thought this stuff was cool but had no real uses for it. I was very limited on disk space back then so I eventually threw it all out. Now I'm wishing I hadn't been so thorough. Does anyone remember this app, know its name, or where I might find it again? It was 98 compatible.

We seem to be getting more coders and additional talent here (with more of the discussions going beyond what I understand). And since this is a wish list, how about the ability to at least partially take advantage of dual or multi core processor or multiple processors? Even if this was limited to being able to dedicate or assign the 2nd processor to a single demanding application.

How about a service emulator that can also run the services of apps that use their own, like VirtualBox or SandBoxie? Either or both would be useful as long as we don't end up turning 98 into a service based OS and end up building XP (with all its vulnerabilities) all over again. I realize that it's come up over and over, but since this is a wish list, how about IPv6 support? In a way, this ties in with the above as we'll need the ability to run an IPv6 compatible firewall (not current full blown security suites).

My largest was a Knoppix Live CD, not sure of the exact size. Ran Star Downloader at night for about a week. Was also using CallWave at the time so there was no problems with missing phone calls. The dialup service I used at the time was fairly fast but would only allow you to remain connected for one hour, after which you had to connect again. With that service, Star Downloader was a life saver.

Thanks for the links. ClockMon works great on 98. It doesn't appear to need Kex. I didn't need to do anything with the driver. Just unzip and go. My CMOS clock has been within a second of the time retrieved by ATS. It has been Windows losing time. Running without Vidalia has lessened the time loss by at least 75%. When I ran µTorrent on its default settings to download a rather large file (approx 600MB) from multiple sources, Tor crashed, according to its log "unknown errors". The Tor traffic was fairly heavy at the time and both apps had established a large number of connections. I don't know exactly how many total connections there were, at least 2 full status screens on Kerio, estimating around 150. 98 was stalling and becoming intermittently unresponsive just before Tor crashed. It appears that the maximum number of allowed connections is set too high, presently at 512. The number of allowed connections for µTorrent definitely needs to be restricted. Other than trial and error, is there a realistic way to determine what the maximum number of connections is that an individual PC can handle? Except for the connections issue from using Tor and µTorrent together, everything else has performed well and been stable with Tor running. After making a full backup of the existing system, I updated Kex to the latest version. Going to see what else this new version makes possible and make sure that it's compatible with the rest of the apps I use. I'm interested to see how well SeaMonkey 2.4 works on my system and how it compares to 2.0.14. Tor is restarted, running with the same settings. I have no immediate tests planned for it other than its effect the long term stability of the PC during normal usage.

So far, Kernel Ex has worked well on a virtual 98SE unit. The present version of SeaMonkey (2.4.1) runs but has a few issues. The "set as default" option doesn't work, asks every time. On the virtual system SeaMonkey will not save or import bookmarks. Is anyone else having this problem or is it a problem with my virtual system? The present version of Kex seems to get along with the apps on my default system, including SSM and Tor. I'm holding off updating SeaMonkey for now. When I tried to add my usual complement of extensions to SeaMonkey on the virtual system, most would work with 2.4, including FlashBlock, Ghostery, and Better Privacy. Hopefully those who maintain the extensions can catch up to this ridiculous update cycle they're using now.

Instead of choosing between 98 and either XP or 2K, dual boot the unit and see which serves your needs better. If you don't keep both, the partition that contained the one you don't keep can be used as a data partition. When I first got this Dell, I planned on getting rid of XP and making it 98 exclusively. I ended up making it dual boot. For my needs, the 98 system consistently out-performs the XP system with a few exceptions, mostly game graphics. Until I was given this Dell as payment for other work, my primary PC was 98FE on its original hardware (with some upgrades). It was a reasonably stable unit that performed better than expected for its specs. I definitely agree with you on the software. I also use IrfanView, a slightly newer version (3.80). For my limited office work needs, Microsoft Works is quite sufficient. Open Office works fine too but is too bloated for my needs. Regarding Tiny firewall, are you using one of the 2X versions, 2.0.15 possibly? That firewall is the predecessor to the one I use, Kerio 2.1.5. They're so similar that each can use rulesets made by the other. The newer firewall you describe as intrusive are for all purposes, security suites. Most of them won't work on a 9X system anyway. Even on systems they're compatible with, I prefer separate, freestanding applications, which usually includes Kerio 2.1.5. Absolutely! With newer hardware having almost unlimited memory and processing power, nobody bothers to write efficient code. I don't think most of them know how any more. 98 can fly and multi-task on hardware that can barely power the new operating systems. I'm at a loss to understand how such gross inefficiency and waste can be called progress. I will never understand why users tolerate or accept this. If an individual or company demanded that you repeatedly prove that you did not steal your home or car, you'd eventually file some kind of complaint or harassment charges against them. When a software or OS vendor does this, why do we tolerate it? For me, this alone is sufficient reason to boycott versions of Windows newer than 2K. On the single XP system that I have, which is legitimate,all of the WGA components have been removed, and the OS is not permitted to connect to Microsoft. If the same standards that are used for applications were applied to operating systems, XP and newer systems would be classified as spyware. On a 9X system, closing the open ports took about 2 minutes and a reboot. Closing all of the open ports on XP is a much bigger job, requiring a lot of disabling and changes to system services. There are entire websites devoted to that subject. The default attack surface of the OS, most of which the average user will never use, is huge when compared to a 9X system. On Win-7, I'm not sure it's possible to actually close all of the open ports, not just block access to them with a 3rd party firewall. I'm convinced that the newer versions of Windows are designed to spy on its users, and that it gets worse with each new version of Windows. I also question the effectiveness of the built in security features that are part of the newer versions of Windows. Look at the recent findings that shows how AppLocker and SRP can be defeated using LoadLibraryEx. Most will call my view paranoid or worse, but IMO, this does not look like a coding error or a bug, but a deliberate bypass of the built in security features. I can't help but wonder how many more of these there are that haven't been "discovered". On a 9X system, a single batch file can remove all of the user activity records, including those stored in the registry. Try to do that on XP. If learning Linux didn't feel like I was completely starting over, I'd probably stop using Windows for everything, save a couple of games. After using DOS for so long, the Linux command line syntax feels all wrong to me, like a foreign language. If the unofficial development for the different 98 upgrades ever stops, especially KernelEx, then I'll have to learn how Linux works. Until then, I'll keep using 98 because it works, it's stable, easy to secure, and above all else, I enjoy working with it.

The answer will depend largely on what you plan to do with your PC. Assuming compatible hardware, 98 will be faster than XP or 2K. With the unofficial upgrades (Kex, RP9, NUSB, etc) 98 can be just as stable and reliable as XP. As long as development continues on these upgrades, compatibility with the common user apps isn't a big problem, unless for some reason you prefer Microsoft applications. 98 is at its best with 3rd party and Open Source applications. Regarding security, there's a bit of a tradeoff here. Most malware isn't targeting and doesn't run properly on 98. Most AVs don't run on 98 anymore either. If you're comfortable with setting up and using a default-deny security package, 98 can be made just as secure as any other version of Windows, maybe more so due to its small attack surface. On the subject of privacy, 98 keeps far fewer records of user activities. Thanks to DOS, they can all be accessed and/or deleted. The biggest downside to 98 is that there's almost no vendor support for both the OS and the installed applications. For a casual or typical user, this is a problem. For a skilled user capable of doing it themselves, it's not an issue. If you want an OS that makes the decisions for you and doesn't require a lot of input from the user, choose XP. If you want an OS that does what you tell it, one that doesn't try to override your decisions, 98 is better.

Updating Kex requires a reboot which I really don't want to do at this time. My PC is in the middle of a long term durability test of Tor as an exit node on 98, which Kex and RP9 made possible. Tempting as it is, I don't want to cut it short. All I wanted to know was if FireFox (which I don't use) and SeaMonkey were affected by the same problem.

When I read your "call for support" thread, I thought that was the end of KernelEx. This is a very pleasant surprise. You just made my day and then some. Maybe this will help with that. 359 downloads of this update already! A quick question. Regarding the problems with newer versions of FireFox, is this the same problem that affects SeaMonkey versions 2.1 and newer? This also has me wondering if the present Tor browser bundle (which contains a modified version of FireFox 8) will run on 98 now.

98 units can be had for next to nothing at yard sales and resale shops. As for running it on newer hardware, PCs with Pentium 4 processors make excellent 98 units. Hardware that's newer than that can be very problematic. This thread will give you a good idea of what newer hardware 98 will run on.

ATS has worked well for me, after I found an easy to contact time server, but I would prefer to eliminate the problem instead of constantly correcting the clock. Is there any realistic way to prevent it? I'm monitoring processor usage via Process Explorer, resource usage with the built in meter, and memory usage with MemLoad 2.0. ATS is checking server time every 10 minutes but is not correcting the system clock. Is there something else I should be monitoring in order to track this down? Is there an app that can read the CMOS clock and correct the system clock? After the last restart, I installed a bandwidth monitor and changed the Tor configuration file to log to file instead of sending the output to Vidalia. For now, I've stopped using Vidalia entirely. With the bandwidth monitor and Tor logging to file, Vidalia doesn't serve much purpose. The relay has been back up for a bit over 4 hours, but the traffic level and number of connections are not yet up to their earlier levels. That will probably take a day or so. So far, the system clock has lost no time. During the previous Tor trials, I restricted the usage of this PC to light tasks like web browsing, light CD work, etc. While I want it to be a stable Tor exit that contributes to the network, I can't limit it to that role. It is my primary PC so I will not be limiting its usage during the current test run. So far, it seems able to run Tor and Virtual PC at the same time, but it's only been a bit over 5 hours since it was restarted.

Is there any realistic way to prevent this or is an app like Atomic Time Sync the only real option? I have seen the system clock lose time when processor usage hits 100% for an extended period but I don't think that's happening here. Most of the time, I see very little processor usage.

I have it. It doesn't supply the running traffic totals I want. At the moment, I'm trying NetMeter 1.1.41 beta from metal-machine. Just installed it moments ago, seems quite good.