disk write error during unattended update

[Chris Solomon]

several times now i have come back after letting my system sit overnight (defragging another drive) only to discover an update to this and one other game has errored out with a disk write error. now, normally this error occurs if there is not enough space or if there is some sort of drive error. the drive in question is an array comprised of a 1Tb WD drive that came with the computer and an Intel Optane 32gb Module. the two pieces of hardware appear as one device in windows device manager as Intel Optane + 932GbHDD. I have unpaired the devices (temporarily) and checked each drive separately. there are no sector errors on either device. there appear to be no file system issues with the drive array either. and there is 192 Gb free on the drive (D:\). I can find no possible cause for the error. file permissions don't seem to be the cause either, as when I click retry, the update completes sucessfully. I am at a loss as to what is going on here. only possible causes *might* be Steam not liking my use of an Optane module to accelerate the HDD I/O. Or, it's some sort of steam client issue.

anyone else have any issues with use of an Optane setup, or with unexplained game update write errors?

With the Optane module accelerating the HDD, the HDD R/W speeds are as fast as that of a slower SSD. windows even recognizes the array as an SSD. but i use Diskkeeper 18 and defrag the drive as an HDD just to keep the HDD portion de-fragged.

[cswiger]

The number of results from https://www.google.com/search?q=optane+disk+write+error+steam suggests yours is not a unique problem.

Maybe ask Intel and include the exact error messages from Event Viewer...?

[Chris Solomon]

that's the thing, there doesn't appear to be an associated event in the windows event log that i could find. and the issue seems to be limited to only satisfactory and subnautica below zero.

[Die Hand Gottes]

If the hard disk is not defective, hopefully also checked, updates need much more memory than just the file that is downloaded.
fiede must be unpacked then also the data from the game must be unpacked this then patched and the update is removed and the game file patched again.

[kimba-rip]

maybe defragg'in and update'in at the same isn't wise.....

[Chris Solomon]

i was only defragging another physical drive. since it's is a friggin huge 5tb in size i left the system running to defrag during the night and at some point an update dropped for satisfactory, whose files reside on the optane drive.

[Chris Solomon]

Originally posted by Die Hand Gottes:

If the hard disk is not defective, hopefully also checked, updates need much more memory than just the file that is downloaded.
fiede must be unpacked then also the data from the game must be unpacked this then patched and the update is removed and the game file patched again.

its not defective. and messages that i might see in the event viewer from disk service of NTFS are not appearing. am refering to an error like this - "The driver detected a controller error on \Device\Harddisk2\DR2" (note, this is absolutely not the optane drive in question. this is an example!)

I typical DO see the messages coming from my 5Tb external drive when it starts to get defragmented, and also when that drive is under heavy load. but the error has not been logged since early august, so it's been a while since it last appeared, and it is not showing up at all for the optane drive array.

the only thing i am seeing is NTFS going through the drives one by one and reporting that they are healthy.

[Chris Solomon]

I am seeing in the event log some pretty unusual error events for TCP however. not sure if its related, or for that matter, what the errors are even saying.

they are:

event ID 4227
TCP/IP failed to establish an outgoing connection because the selected local endpoint was recently used to connect to the same remote endpoint. This error typically occurs when outgoing connections are opened and closed at a high rate, causing all available local ports to be used and forcing TCP/IP to reuse a local port for an outgoing connection. To minimize the risk of data corruption, the TCP/IP standard requires a minimum time period to elapse between successive connections from a given local endpoint to a given remote endpoint.

event 4226
A request to allocate an ephemeral port number from the global UDP port space has failed due to all such ports being in use.

Event ID 4231
A request to allocate an ephemeral port number from the global TCP port space has failed due to all such ports being in use.

[Chris Solomon]

if the there is some sort if issue that is causing the steam downloads not to complete properly this might potentially cause a write error if there is some sort of issue with the file being worked on. this would only be if steam uses delta patching to update the games though. not sure if steam does that, or what. the write error it self would occur when the patched file was being transfered from the ..\steamapps\downloads\ directory to the games directory. if there was some sort of file mismtch between the orginal unpatched, and the new (supposedly) patched file. in this case the error is occuring becuase steam is written well enough that it will try not to corrupt our game files with a badly downloaded patched file.

[Huperspace]

more likely an AV blocking access to a file, so steam can't overwrite it and throw this error.
excepting the game folder from AV with a reboot should fix it.

[Die Hand Gottes]

Originally posted by Huperspace:

more likely an AV blocking access to a file, so steam can't overwrite it and throw this error.
excepting the game folder from AV with a reboot should fix it.

Correct the antivirus scanners are often over motivated disabling and testing if the update goes through.
I already know why I don't use them anymore.

[cswiger]

Originally posted by Chris Solomon:

I am seeing in the event log some pretty unusual error events for TCP however. not sure if its related, or for that matter, what the errors are even saying.

These aren't related to your Optane hybrid disk write issues.

Either you are being heavily port scanned, or you are running something unusual like a badly configured Bittorrent client, or you might even have malware which is performing a DDOS attack against someone else.

[Chris Solomon]

i don't have the bandwidth for bittorrent, sadly. as for the being port scanned.... idk, how would i tell? i certainly don't have any malware on my system. my bi-monthly system upkeep would have discovered that if i did. as for AV atm I am using the version of windows defender that comes with my windows 365 subscription. i will see if adding an exception for my steam directory helps

[cswiger]

You run a network sniffer like tcpdump, or maybe look at your Internet router/firewall if it has decent capabilities.

[Chris Solomon]

um... it does have some logging capabilities. I assume anything like this would be under the DDOS protection logs. only thing is that particular log is notorious for bieng filled with false positives. if i believed only the router logs, my ISP's servers are constantly attacking me.