Is it the status quo for this project to produce a huge number (43% of WUs) of "errors while computing???"

<core_client_version>7.9.3</core_client_version>
<![CDATA[
<message>
process exited with code 195 (0xc3, -61)</message>
<stderr_txt>
05:33:36 (31482): wrapper: starting v1.3.9.7
05:33:36 (31482): device: 28 x Intel(R) Xeon(R) CPU E5-2683 v3 @ 2.00GHz [Family 6 Model 63 Stepping 2] (3.100020 GFLOPS)
05:33:36 (31482): checkpoint interval: 70 min (task 12600 GFLOPS, 68 min)
05:33:36 (31482): wrapper: running dnetc517-linux-x86-elf-uclibc (-ini dnetc.ini -runoffline -multiok=1) - attempt 1/10

dnetc v2.9108-517-CFR-10032922 for Linux (Linux 5.3.0-62-generic).
Using email address (distributed.net ID) '<censored> '

[Jul 16 12:33:36 UTC] Open failed for './in.r72'
Unable to obtain directory information.
[Jul 16 12:33:36 UTC] Buffer seek/read error. Partial packet discarded.
[Jul 16 12:33:36 UTC] Shutdown - packet limit exceeded.
[Jul 16 12:33:36 UTC] Open failed for './in.r72'
Unable to obtain directory information.
[Jul 16 12:33:36 UTC] Open failed for './in.r72'
Unable to obtain directory information.
[Jul 16 12:33:36 UTC] *Break* Shutting down...
[Jul 16 12:33:36 UTC] Shutdown complete.
05:33:37 (31482): app exit status: 0xfe
05:33:37 (31482): called boinc_finish

</stderr_txt>
]]>

In the last 20 minutes the failures have already increased to 46%.
There are 19 Invalids but every time I click on Invalid on the Tasks page I get Service Unavailable.

It's clear that CPU WUs are a disaster. I disabled CPU work and returned all the ones I have.