Raphael Raph Carvalho

thanks for the detailed report @IsaacJT. I'll work on that ASAP

@JulianLai what the following command returns: "cat /proc/kallsyms | grep sys_call_table"? could you please also share the backtrace from gdb?

Hi @sidro. I'll find a way to fall back to a different approach when those files aren't usable. Thanks!

@sidro: do you have /boot mounted? could you please try running the program as root instead?

@JulianLai: this commit 9b740460c9f36e0d0178ab8d646a9400138cadaf may help you, please pull changes and try again

thanks for the report @sergiy-kozak. i'll work on fixing that ASAP.

@sergiy-kozak could u pls consider running it as root? that will give me direction

A local test, using same commit as the test, shows resharding on a shard S is still serialized and space of input files is released after the job has completed....

> > I guess we can litter them with maybe_yield() as the first statement. The alternative is to add yields in coroutine::parallel_for_each(), though that penalizes the common use case of...

> Upload UUID = 94a706fe-0906-4f16-8dd8-f3af4a578a60 _Installation details_ Scylla version : 5.0.1 Cluster size: 9 OS (RHEL/CentOS/Ubuntu/AWS AMI): Ubuntu > > I stopped Scylla and deleted some corrupt files then restarted...