<div dir="auto">Do you have any suggestions about software that will check for data corruption by continuously stress a storage ? (Like creating files, read them back, move around, renanes, deletes and so on)</div><div class="gmail_extra"><br><div class="gmail_quote">Il 12 ott 2017 11:15 AM, "Robert Altnoeder" <<a href="mailto:robert.altnoeder@linbit.com">robert.altnoeder@linbit.com</a>> ha scritto:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">On 10/11/2017 11:30 PM, Eric Robinson wrote:<br>
> The TrimTester program consists of three parts. The main executable<br>
> (TrimTester) just writes loads of data to the drive and tests for file<br>
> corruption. My C++ consultant says, "It writes sequential numbers<br>
> wrapped at 256, spanning multiple files. It checks previously written<br>
> files, and if the file data is all zeroes, it is considered to be<br>
> corrupted."<br>
Are you referring to this program?<br>
<a href="https://github.com/algolia/trimtester/blob/master/trimtester.cpp" rel="noreferrer" target="_blank">https://github.com/algolia/<wbr>trimtester/blob/master/<wbr>trimtester.cpp</a><br>
<br>
One thing that I can tell you right away is that this program does not<br>
appear to be very trustworthy, because it may malfunction due to the use<br>
of incorrect datatypes for the purpose - apparently, it is attempting to<br>
memory-map quite large files (~ 70 GiB) and check using a byte-indexed<br>
offset declared as type 'unsigned', which is commonly only 32 bits wide,<br>
and therefore inadequate for the byte-wise indexing of anything that is<br>
larger than 4 GiB.<br>
<br>
While this indicates that the program might miss actual corruption, so<br>
far I have not found any definitive proof that the program will generate<br>
false positives (however, I did not check the program in much detail,<br>
judging by the overall quality I would not be surprised if it did), so<br>
we should still continue investigating.<br>
<br>
I would certainly recommend to double-check by running some other<br>
software to check for data corruption issues to ensure that the problem<br>
is not malfunctioning test software.<br>
<br>
br,<br>
--<br>
Robert Altnoeder<br>
<a href="tel:%2B43%201%20817%2082%2092%200" value="+43181782920">+43 1 817 82 92 0</a><br>
<a href="mailto:robert.altnoeder@linbit.com">robert.altnoeder@linbit.com</a><br>
<br>
LINBIT | Keeping The Digital World Running<br>
DRBD - Corosync - Pacemaker<br>
f / t / in / g+<br>
<br>
DRBD® and LINBIT® are registered trademarks of LINBIT, Austria.<br>
______________________________<wbr>_________________<br>
drbd-user mailing list<br>
<a href="mailto:drbd-user@lists.linbit.com">drbd-user@lists.linbit.com</a><br>
<a href="http://lists.linbit.com/mailman/listinfo/drbd-user" rel="noreferrer" target="_blank">http://lists.linbit.com/<wbr>mailman/listinfo/drbd-user</a><br>
</blockquote></div></div>