Possible SSD Issues

[Antec]

Hello,

Where we are ?

I do not know exactly with which versions 3.5.1. xxxx, but about the date of your posts ... I burned the SSD Kingston of a PC (the personal one that I take to work) ... during the personalized scan, he got stuck on the analysis of memory ... Hard Disk Sentinel alerted me, the temperature went up to 62 ° ... I stopped the scan ... but too late the SSD had suffered too much, it melted some time later .. (Arf, I was at work, and not watching the PC ... but the job)

And it still happens from time to time, that the scan block on the analysis of the memory ... (now I pay attention during the scans ...!)

Also, where are you?

FJ

 

[exile360]

3.5.1 is still the latest version.  The fix dcollins referenced above is still in testing and should be included in the next Malwarebytes 3 release, however I do not know when it is expected to become available as it depends on how testing and development go and if any further bugs or issues are identified which must be fixed and tested in the new build.

[dcollins]

As mentioned, this issue should be fixed in our next release which I believe is slated quite soon.

As for the SSD, your issue doesn't quite align with the symptoms of this thread since this just leaves zombie processes open on your computer using very small amounts of RAM that eventually add up to quite a lot. If you're having scanning issues, I would create a new thread and provide the logs requested in the top of this thread.

[Antec]

On 5/31/2018 at 1:46 AM, dcollins said:

As mentioned, this issue should be fixed in our next release which I believe is slated quite soon.

As for the SSD, your issue doesn't quite align with the symptoms of this thread since this just leaves zombie processes open on your computer using very small amounts of RAM that eventually add up to quite a lot. If you're having scanning issues, I would create a new thread and provide the logs requested in the top of this thread.

 

Hell

to drop for the SSD ... it's a consequence, and not a repetition. At the limit of my fault ... you must always have your eyes everywhere

But the blocking problem on the memory scan is repetitive. But not systematic.  And whatever the PC.
Latest version also 3.5.1.2522 - 1.0.374 - 1.0.5366 at the moment ...
Analysis on Ram still running at 6.30mn. Stop and restart, the complete scan (Threats) lasted 6mn ... analysis on Ram, 2mn ...

On other PCs, it's only a few seconds (except when it blocks lol)

 

I do not think I'm alone with this problem, especially since it's been around since well before version 2522

Sorry, hardly any time to make logs right now ... and you know what's going on

Good luck

 

 

[dcollins]

Different machines can expect to have different scan times, since they'll have different files on the drive and different running processes. I know you stated you're busy, but since no one else has reported the issues you're describing, we will need logs to help us understand what's going on.

 

1. Download the Malwarebytes Support Tool
2. Once the file is downloaded, open your Downloads folder/location of the downloaded file
3. Double-click mb-support-X.X.X.XXXX.exe to run the program
   • You may be prompted by User Account Control (UAC) to allow changes to be made to your computer. Click Yes to consent.
4. Place a checkmark next to Accept License Agreement and click Next
5. You will be presented with a page stating, "Welcome to the Malwarebytes Support Tool!"
6. Click the Advanced Options link
   
7. Click the Gather Logs button
   
8. A progress bar will appear and the program will proceed to gather troubleshooting information from your computer
9. Upon completion, click OK
10. A file named mbst-grab-results.zip will be saved to your Desktop
11. Please attach the file in your next reply. Click "Reveal Hidden Contents" below for details on how to attach a file:

[siliconman01]

My experience with MB 3.5 as it pertains to SSD temperature rise during manual MB scans is that the SSD temperature rise will be significant and warrants efforts to moderate.  The same applies to the CPU temperatures.  On my primary gaming system, I use a Samsung 970 Pro SSD.  I had to add a heat sink with a fan to cool the SS 970 Pro.  Without the heat sink, a manual MB scan would raise the SSD temperature from 38°C to 58-60°C.  The 970 Pro does have a built-in temperature setting that throttles back the SSD active automatically if the temperature gets too high.  With the heat sink installed, the SS 970 temperature increases only up to 47-48°C during an MB manual scan.  

 

[plb4333]

siliconman01,

Just curious, didn't this setting help any?

 

[siliconman01]

12 hours ago, plb4333 said:

siliconman01,

Just curious, didn't this setting help any?

 

You know, I too thought it would help; however, it does not...at least on the Samsung 970 Pro.  The temperature rise is just as dramatic.  My testing was done for this by just changing the setting and then letting it scan without me doing anything else on the machine during the scan.  

[plb4333]

That's really surprising. When I think of SSD, to me they should be able to keep up without the overheating. I'm glad to know this, Thank you

[plb4333]

I had just seen where the 970 PRO SSD is rated for 0-70 deg. C.

[exile360]

Also keep in mind that throttling also occurs automatically after a certain number of seconds of sequential reading or writing of data from/to the drive.  I don't know the specs for the 970, but I've got a 960 PRO 1TB and it's rated at 130 seconds for reads and 160 seconds for writes.  I don't know the throttling temps for the 970, but the 960 PRO is rated at up to 75 degrees C and I would assume the 970 would be higher, though if Samsung says it's 70 degrees (which is the same as the older 950 series) then I guess I'd have to take their word for it.

I also found out that the stickers on the 960s and 970s contain copper to operate as a heatsink and that the controller on the 970 has a nickel plated cover/coating for the same purpose (since the controller is the component that generates the most heat and tends to be the most negatively affected by it).

I was also doing a bit of research on possible third party heatsinks for NVMe SSDs and I found a few cheap solutions from Amazon that might work (I'd need something particularly low profile since I have a laptop) so I might end up purchasing one if heat becomes an issue, though so far in my testing my temps have been pretty stable, usually within the 50C~60C range.  I just performed a Threat scan and my drive temp maxed out at 55C during the scan (7x64 SP1, fully patched) with the total scan time taking 2 minutes and 20 seconds with the filesystem portion only occupying the last minute or so of the scan (not counting the rootkit scan at the beginning prior to the memory scan phase which went by pretty fast, though technically it is scanning portions of the disk during this phase, but the temp never went up above 50C during those phases).  During the scan it checked a total of 232,346 objects.  The most notable lag was, as usual, during several portions of the memory scan phase which suffered from noticeable pauses at several points during which it didn't appear to be making any progress until it finally moved on (it typically does this around 4 or so times during each scan) though high CPU usage was fairly consistent throughout, including during these 'pauses'/hesitations (I assume they have something to do with heuristics and/or optimization of the later phases of the scan, and/or possibly the anomalous threat detection component since it relies on cloud analysis for checking of unknown objects).

[siliconman01]

Well, some of my heating issue is centered around overclocking.  I am using an i7-6700k CPU overclocked to 4.6 mhz with a water cooler on the CPU.  And I have 64 gigabytes RAM which provides all sorts of room for things to execute without shuffling.  When I do a manual scan (including rootkit) with MB 3.5.1, it scans roughly 336,000 items in 11 min, 7 sec.  I also get a 2-2.5 minute "pause/delay" when it reaches the end of the C:\Windows\Winsxs folder.  After the delay period, it takes off again within the Users folder and completes the scan after performing the heuristic scan. (I haven't been able to determine what MB is doing during this delay but the CPU power drops off significantly during that period.)   With the air cooler heat sink on the SS 970 Pro, it maxes out at 46-47C.  Following the scan, it cools back down to an idle temp of 36-38C.  

My personal preference/experience is that keeping components cooler prolongs the life span and also improves system performance.  For example, I see a 2-3% performance improvement by using MSIAfterburner to control the fan on my overclocked Nvidia GeForce 980 Titan graphics card.  Of course one could easily argue that "overclocking" and "prolongs the life span" is a counter intuitive statement.

[exile360]

Are you doing any undervolting on your CPU?  That can be a good way to reduce heat when overclocking if your CPU can handle it (the better chips which come from closer to the center of the die tend to allow for more undervolting than the lower quality ones that come from further out, but getting a "good" one or a "bad" one is totally luck of the draw as Intel/other chip makers do not label those parts differently).

Are you doing Custom scans?  I ask because 11 minutes seems really long to me.  My Threat scans are typically around 2 minutes or less, even with my older 4910MQ CPU which isn't as fast as your 6700K.

[siliconman01]

Not doing any undervolting on my CPU.  I'll read up on that because I am not familiar with that tactic.  Thanks for the tip.  I use an ASUS ROG Maximus VIII Extreme motherboard.  The BIOS settings can be a bit difficult to understand unless the user is a computer tech whiz....which I am definitely not.  

Yes, I am doing a Custom Scan.  A threat scan only takes about 1 min 7 sec.

[exile360]

You should check out Intel® Extreme Tuning Utility (Intel® XTU).  It allows you to overclock by manipulating BIOS settings from within Windows, and the changes you make are reflected in your BIOS after you reboot.  It allows you to store multiple profiles, including the system defaults in case there's an issue, and it automatically undoes any changes you make the moment the system crashes due to unstable settings so there's never a need to mess with jumpers on the board etc.

Most of the settings are easy to understand and you can find tons of guides and advice online on how to use it and what each setting is and does and whether or not you should mess with it.

[siliconman01]

Thanks much for the info.  I will definitely look into using XTU.

 

[exile360]

You're welcome, it's a great tool.  It even lets you do benchmarks and compare them with others online if you choose.  It even shows you details on voltage, current and temperature so you can see when the CPU is being throttled for any reason which helps with making sure your OC settings are stable/make sense.