cc32d9
Had this error occurring again few days ago, and that was the only node where I forgot to put `read-mode=head`. It was isolated from speculative transactions though: ``` p2p-accept-transactions =...
there's at least one occurrence of this error with `read-mode=head`. 2.0.13 (WAX) could not write to state history and stopped.
One more occasion, and this time it's not at head block: I was resynching a Telos ship node from genesis, and it stopped about halfway through the history (last block...
attached, is the log from the Telos node (log level is set to error). The node started from near the genesis on Nov 12 , then stopped on Nov 14...
the error is quite definitely related to how busy the storage is for ship files. Seems like the write operation times out too early.
After running overnight, the process occupies 15GB RAM. It looks like there's no control over memory recycling, or maybe it's a memory leak.
the writer in question: https://github.com/EOSChronicleProject/chronos/blob/main/writer/exp_chronos_plugin.cpp ``` cass_cluster_set_local_port_range(cluster, 49152, 65535); cass_cluster_set_core_connections_per_host(cluster, scylla_conn_per_host); cass_cluster_set_request_timeout(cluster, 100000); cass_cluster_set_num_threads_io(cluster, scylla_io_threads); cass_cluster_set_queue_size_io(cluster, 1048576); ``` scylla_conn_per_host is set to 1, and scylla_io_threads=4.
``` [48556796.397656] chronos-writer[1981192]: segfault at 8 ip 00007f60800c5af8 sp 00007f5c777fa9d0 error 4 in libscylla-cpp-driver.so.2.16.2-b[7f607ffd2000+293000] [48556796.397677] Code: 01 00 00 4c 8d 2c d0 0f 85 4e 02 00 00 49...
@jul-stas ^^
well in the meanwhile I dropped the idea of supporting ZMQ, and now using websockets only. But the reproduction case should be quite easy to build, just send to ASMQ...