Nick Vatamaniuc

Curious if you have a chance to test, does this happen with both the default Spidermonkey and the QuickJS engines? You can toggle the config with https://docs.couchdb.org/en/stable/config/couchdb.html#couchdb/js_engine and restart and...

Thanks for trying it out @StephanU. It points to an issue with the spidermonkey process only then. If you want you can keep it on QuickJS. We converted almost all...

As a bit of a background, I was inspecting one of my executables which embeds QuickJS and noticed in the symbols listing with `nm -u` it was showing a bunch...

Just a wild guess, but since we see this in OTP 25 but not in OTP 24 wonder if it might be related to https://github.com/erlang/otp/pull/5195 (Further optimize off-heap traversal during...

Not sure if's helpful or not but another core dump has a similar cycle just a bit smaller, with only two items: ``` (gdb) bt #0 erts_deref_node_entry (term=139983200562610, np=0x7f511fb78c00) at...

Another interesting thing I noticed is that all the processes which attempt to clean the off-heap list just one particular process in Apache CouchDB: https://github.com/apache/couchdb/blob/main/src/fabric/src/fabric_db_update_listener.erl#L39. ``` (gdb) etp-stacktrace *(Process*)(0x7fce173f1f98) %%%...

Some debugging info about the schdulers ``` (gdb) etp-schedulers --- Scheduler 1 --- IX: 0 CPU Binding: unbound Aux work Flags: delayed-aw-wakeup delayed-dealloc delayed-dealloc-thr-prgr later-op canceled-timers misc Sleep Info Flags:...

Thanks for taking a look, @sverker. Here are the nifs/drivers we have: ``` (gdb) etp 225227 exxhash. (gdb) p *((struct tainted_module_t*)first_taint.counter)->next $2 = {next = 0x7fab2175c8d8, module_atom = 254923} (gdb)...

I am also now able to reproduce the issue on a smaller, isolated test cluster. I compiled a debug emulator and ran it with the load that usually reproduces the...

Thank you, I'll compile it and give it a try.