Cav's Place

2018’s last run was yesterday, the time being 48:23, with sector times of 4:15, 5:06, 5:54, 4:32, 5:05, 6:00, 4:37, 5:08, 5:59 and 1:47, making for lap times of 15:15, 15:37 and 15:44. Had set the alarm at noon, but woke up the second time just before 11 AM and decided to get up, eventually going out the door just after 12:45 PM. The weather was good, with a reported temperature of 5°C but quite sunny and with pretty much no wind, but there were a number of people to find my way around. Plenty of runners too, but those aren’t a problem, and one likely even helped, as he overtook me on sector three of lap two and ended up quite far ahead, but obviously couldn’t maintain the pace and I caught up to him and overtook him again on sector three of lap three, that little “race” likely making me be a little faster than I’d have been otherwise, though by then I was already pushing hard. Did feel rather out of breath most of the time, and that was the limiting factor starting even after the first lap, and definitely from sector two of lap three, when I started giving it everything, since I was in real danger of failing to stay under 48:30, so I breathed out through my mouth plenty of times and on parts of the final sector I breathed in like that too, though at the end I saw that I didn’t need to do that, since it likely helped me gain two or three seconds, not seven. The knee was really fine though, so that was nice, though another issue was that I felt like I needed to take a crap almost from the beginning, despite having managed to get rid of a bit before leaving.

After the run, since I had taken a bit of money with me, I went to the Carrefour from that area and got myself a couple of things, though it was awkward with the one I had to ask for, being sold by weight like that. I said 200 grams and the employee put 238, then seemed bothered when I said that’s too much, saying it’s hardly anything over what I had asked for, making a show of taking just a tiny bit away, asking quite angrily if that was how much it was over… Well, no, not just that much, since what she took away was just 12 grams, but it was good enough for the money I had left. Actually, even the initial quantity would have been good enough, since I found a coin at checkout and then the guy from the shop where I charged my phone card, with just 1 EUR, only to activate it again in case it’ll be needed these days, rounded down the amount he asked for, but I couldn’t have known that then and that initial amount would have put me over what I had by a tiny amount. Plus, it was 19% over what I had asked for, which is a large margin.
Either way, got back just in time for ski jumping, and to catch dad before he left, since he had said he’ll leave around 4 PM but apparently decided to go at 3 PM… And a few minutes later came back for something, and I reminded him of a couple of other things I noticed he had forgotten too. But after that, not counting the cats, I was alone for New Year’s, which is at least less bad than being alone with someone else around. And otherwise these days just mess up my routines, as I wouldn’t care for them one way or the other. The New Year is hard to ignore though, and it did bother me a bit that I didn’t have what I consider typical food for that night around, since he only brought some of those things, and quite little at that, when he came back today. But I still stuffed myself with a few things that were left and the usual stuff I make, even managing to do it all early, finishing eating just before midnight… And later I decided to start another game at the start of the year, installing Jotun and completing the first area before going to bed.

Since I’m on the topic of games, the day before I won Oblivion on another truly massive giveaway on the GOG.com forums, organized by the same guy who does these massive giveaways in which I had already won first The Witcher 3, back in spring, and then, in September, a free choice of a game, which I used for Lords of Xulima, plus that he also donated the copy of Divinity: Original Sin I requested and received from the Community Giveaway, in October. And this time he wanted to make sure that everyone who entered received something, at first adding more codes, including for Oblivion, which hadn’t been on the initial list, before the deadline and telling people to feel free to edit their requests to add anything new they want to enter for, which I did, then actually adding quite a number of additional codes at the time of the draw, for games requested by many, then asking for those who didn’t win anything to select something out of the codes that nobody, or not enough, had entered for, and then apparently offering something to those still left out, until he said that everyone who entered and whose chat settings didn’t block him from contacting them had received something, and there were messages from those who had mentioned not winning anything saying they had indeed been surprised to receive a code anyway. But I seem to have won the first code for Oblivion, not anything added at the time of the draw or later, so at least I don’t have even more reasons to feel guilty and embarrassed. Already do anyway, considering all I received from him in so short a time…

That evening was nice, but the first part of the day was rather annoying due to Micky. I had opened the window instead of turning on the exhaust hood while making tea, and despite closing the door she opened it, came into the kitchen and then jumped on the window, refusing to come down when I wanted to close it. It’s possible that she wanted to get on the cupboard, but that doesn’t go that far anymore, due to the new exhaust hood, and I was afraid she’ll fall since she was getting agitated if I tried to get her away from there, clung on to the window, harder when I tried to pick her up and get her down, which meant I couldn’t do so when I had to stretch so much to reach. So I eventually got on a chair to pick her up directly, but that allowed her to get on my shoulders, as she knows she can do with dad, having him carry her wherever she wants, and I couldn’t get her off, since when I got down and tried to pry her off she just clung on harder and sounded quite annoyed. Tried to avoid pulling her off directly, carefully pulling each claw out of my shirt until I could finally get her off, but that still left several holes in that shirt, which had been quite whole so far.
On the other hand, and back to games, Sunday I also managed to get myself to write the review for Age of Wonders: Shadow Magic, submitting it on MobyGames too just before midnight, after going over it again and making a few changes. Still found one more word I had to add when I went over it one more time, that night, but I can edit things there, so just edited it in the queue, without pulling and resubmitting it. So I did manage to write it before the end of the year after all… And then even added one more post yesterday, albeit a quick one, getting the number of bands included in “New Finds” posts during one year to at least 15 for the first time.

And now I have to get to the issues mentioned at the start of the previous personal post, which started on the night of December 26 to 27, when I got back on the computer after eating, did a few things and then noticed that some things which should have been blocked by Emsisoft Anti-Malware (EAM), as privacy risks, were loading on sites, and this was how I was used to notice that the bug which caused its surf protection and behavior blocker modules to turn off invisibly had triggered again. It hadn’t happened in months, and while I had identified things which caused it to happen and really kept doing my best to avoid doing them, I doubted I had managed to completely avoid every single instance in so long, so even thought that they had fixed it at some point, even though the support guy I talk to had told me that they hadn’t worked on fixing anything about the old surf protection due to developing an entirely new surf protection module, which wasn’t included in the version I had, being on the “delayed” branch. But, of course, my first reaction at that moment was to assume I had indeed been wrong in thinking they had fixed it, that it simply happened again, and that I just had to fix it the same way I did before… Not that I really remembered how at first, so just shut it down and then started it again, before checking the post to see that the solution was to disable and then enable the protection.
Well, that’s where things got bad, because doing that completely froze my computer. Could still move the mouse, it seemed to react normally to hovering the cursor over something, the music was playing, but at first the EAM window was unresponsive, and then I noticed that I couldn’t actually do anything, couldn’t even start Task Manager, trying to close the browser made it appear to close but the music continued to play, trying to open a file from the desktop just froze the desktop as well, pressing Ctrl+Alt+Del did nothing… So, after waiting for a bit, I was forced to press the reset button, at first restarting in Safe Mode and trying to run EAM from there, which at first resulted in a message stating that it’s running without active protection in Safe Mode, so to just scan and then reboot normally, but then it crashed, with a crash message that time. So rebooted normally, and EAM was disabled when I did. That was normal, I guess, so I tried to turn protection back on gradually, surf protection appearing to enable, but then when I enabled file guard as well it didn’t show up as enabled, and a bit later EAM just shut down. Rebooted again, it was still disabled, so I tried to enable all components again, and that led to another system freeze, though at the time I had Task Manager already open and could see some activity, just that at first I couldn’t start or stop anything, and eventually CPU use dropped to zero, so BOINC projects were unable to run anymore either. There was some occasional, slight, CPU use showing up, but I’m not sure if it was only from EAM processes or others as well. And I definitely couldn’t reboot normally.
One thing is that, since the bug which required disabling and enabling protection hadn’t triggered in months, I also hadn’t done so on that version, so had no way of knowing whether there was a bug in that version that prevented protection from being enabled again somehow. And, either way, if I’d have been certain that EAM was the only problem, I’d have uninstalled and reinstalled it, which I ended up forced to do later anyway, even if I kept trying to avoid that. But the thing is that I feared I had exposed myself to some malware around that time, possibly even just before noticing the problem, and that made me panic. The risk should have been very low, if it wouldn’t have been I wouldn’t have done it, but I had been aware of at least a theoretical possibility at the time and having security software disable itself and not allow protection to be enabled again right after that made me assume something bad had happened, which in turn led me to restore the system from a backup done a few days before. It’s the first time I do this and it’s a backup image made with Windows Backup, not some other software, but I didn’t sit to think too much about it. Did hope I’ll be able to restore it on another partition, in fact on another disk, but I didn’t remove the one the system was, and still is, on and it automatically restored itself there, so it overwrote everything and there’s no way to get back to what it was just then.
The system started fine after that, with EAM also enabled and appearing to update just fine, since the time of the last update was obviously before I had created that image. However, one thing I noticed right away was the fact that Windows Update forgot everything, at first noticing that it said it had never checked for or installed updates, and then that the update history was completely gone, though I still see everything if I check installed updates. Also, the search indexes were damaged or missing, so I got a bunch of errors from that in logs, but there were no more of those after that time. And between first posting this and going over it again I also noticed that the maximum disk space for system restore points was set to zero, so I wonder what else I still haven’t noticed. The loss of the update history does bother me though, since it was a simpler and much faster way to check that list, and I wonder if it’s not a sign of a more serious problem somewhere. Plus, other than the EAM crash dumps, since I have those enabled, there’s obviously no way to get logs or system information from before restoring it, and therefore no way to actually know for sure that there was no malware.
It’s extremely unlikely that there was any though, and that’s because the “problem” remained. When I checked again, it still appeared as if EAM’s surf protection didn’t work, which I couldn’t explain. I mean, I had restored a system image made days before the problem appeared, how could it still be there? And then I even switched to the “stable” branch, it seemed to update fine to the current version, yet the problem still didn’t go away… Or at least I thought it was a problem and that it wasn’t going away, because I kept checking the same way and it wasn’t blocking what it should have been, and I definitely didn’t expect a significant change introduced by a simple definitions update, around midnight on the night of December 26 to 27 in Europe, so at a time when I doubt anybody would expect important changes.

With the system at least seeming stable for the moment and no obvious signs of malware otherwise, I somehow managed to go to bed that morning, but just after 7:30 AM I had to wake up and take a shit. I had almost crapped myself due to panicking over what had happened, but decided to just try to go to sleep, and obviously my body decided that it wasn’t going to take that for long. So I sorted that matter out, then turned the monitor on just at 8 AM, saw no issues with the backup script, ran some of the things I was told to run after sending a support message before going to bed, then went back to bed and eventually got back up a bit after 1:30 PM. At that point, since I had a different EAM version and also knew that restoring from backup worked, so if all else failed I could just do that again, I risked disabling protection and enabling it again one more time, but then immediately pressed Ctrl+Alt+Del, waited some ten seconds and rebooted. If it would have wanted to freeze again, it would have likely done so, even more so since the desktop obviously showed up again for a moment before rebooting, so I took it as a good sign that I could reboot normally, though I definitely still wouldn’t care to try disabling and enabling protection again if I can help it in any way.
Well, the good signs only lasted until the system booted back up, since it again waited for a while before attempting to show the desktop and then I was greeted by the same black screen and conhost.exe crash that first happened in August. I still don’t know why it started happening then, but that assumption that it has something to do with .NET updates has since been proven wrong and the one that it has something to do with how the Intel video driver loads may be getting some evidence in its favor. Either way, while it had already been proven to happen every time if I tried to log off and back on, rebooting again after it happened once had always fixed it, so I took it as something relatively normal… Until it happened again after I rebooted one more time, and then several more times. There was just no way to see the desktop anymore, nothing worked, so I panicked all over again at first, but then realized that the only thing that had changed had been the version of EAM, so rebooted again in Safe Mode, which worked, uninstalled it, rebooted normally, no longer experiencing that issue, and then installed it again, which again seemed to work just fine… With the exception of the fact that it still wasn’t blocking what it should have blocked.

After exchanging a number of messages and sending all I could send to the support guy, I then just took it easy, trying to make sure I won’t do anything that risked triggering another problem… And then on December 29 I noticed that there hadn’t been a problem in the first place, not until I disabled protection thinking that there had been one at least! Noticed that because I checked the logs after waking up that day and saw that it had blocked a “privacy risk” the night before, so I started testing more thoroughly, and only then thought to actually check the surf protection list, finally noticing that the “privacy risk” category had been almost completely cleared, only 13 entries being left, including the test one. The support guy wasn’t aware of this having been done either, saying he’ll inquire about it and eventually revealing that the category had caused trouble and there had been talks to remove it for some time, and I’m supposedly the only user he knows of to use it in a proper way, but he definitely hadn’t been aware of any decision having been made, so it was either a mistake or somebody just deciding to do it without telling anyone, so something really troubling and which shouldn’t happen, especially in a security company, either way. He never did tell me which of the two it ended up being, just saying that he also doesn’t know why those few entries were left instead of all being removed.
So I at least knew that my security software did function as intended, and I had already confirmed before that point that the behavior blocker worked as well. However, further tests revealed that their new surf protection module doesn’t exactly work in Internet Explorer. It blocks the test entries if I try to go to them directly, but that’s about it. Tried something that’s listed as a phishing host but which I know is, in itself, safe, and it loaded just fine, and then made a few custom entries, telling it to block them, and it failed to do so every time, though they were blocked, at least under that initial scenario, in both Vivaldi and Firefox. However, while I guess I shouldn’t go into details just yet, under different testing conditions, and I’m talking of very simple and common scenarios, it completely failed to block anything, in any browser. I was also told that the new surf protection module won’t block a site if it has successfully loaded in the current session, so proper testing would require logging off and back on or rebooting, which for obvious reasons I wasn’t about to try, but it wasn’t necessary anyway, since the support guy was able to reproduce all I had reported and said he immediately passed on the information and they are looking into it, because it is indeed a huge vulnerability… Which I noticed right away but which their testers hadn’t picked up on in all this time. Not that it’d really matter to me even if it would be fixed, if it still won’t work in what remains my browser of choice, despite all the shit I get over that. Plus that their “privacy risks” category was a selling point from my point of view, considering how I made use of it, and that quite clearly won’t come back.

So, at this point, I’m pretty much waiting for results and hoping I’ll have solutions in time. There are those crash dumps made before restoring from backup, which should give them an idea about what happened that first time, but I got nothing back about them yet, nor about that conhost.exe crash… And this is the more important matter at this point, because even if there won’t be a power failure or some other problem that will require a reboot before then, there will be Windows updates next week, and if that is indeed, as it appears, caused in some way by EAM and I need to uninstall it to be able to get to the desktop and then reinstall it later, that’s definitely not a sustainable solution. In addition, there’s the matter of a new version likely to be released now and me not wanting to get switched to it, with who knows what other effects, so I’ll need to switch back to the “delayed” branch, but currently that would just put me back on the version that froze my system. The support guy said the “delayed” branch should get switched to what is currently the “stable” one around the same time the new version will be released to “stable”, but that will mean always checking, stopping updates when I’ll see the post about the new version and then relying on him to let me know when I can switch back to “delayed” and update again, since the “delayed” version changes aren’t listed anywhere. So plenty of reasons to stress and worry, as if I didn’t always worry about all sorts of things anyway.