Hi Litecoiners, I am running a node on a cheap VPS mainly to learn. So I just wanted to know if there is the possibility to limit memory allocation to a process, so to not run the risk to trigger the oom-killer? I am looking for a variant so that the process itself is aware how much memory it should use… (Of course, one could disable the oom-killer or give the litecoin process a higher priority*, not running the qt-GUI, but wouldn’t it be elegant if the litecoin process itself was aware about how much memory to allocate at max?)
grep "Killed process" /var/log/syslog kernel: Out of memory: Killed process 1696 (litecoin-qt) total-vm:5752692kB, anon-rss:2163576kB, file-rss:3464kB, shmem-rss:2688kB, UID:1000 pgtables:12264kB oom_score_adj:0
additional info:
litecoin 0.21.3, qt-GUI variant
Adjust /proc/<PID>/oom_score_adj to -1000 (which automatically takes oom_adj to -17 and oom_score to 0). *
OK, I finally switched from litecoin-qt to litecoin-cli … hate it
3 GB RAM wasn’t enough for Ubuntu24 Server + XRDP + Litecoin-qt, even after giving Litecoin-qt high priority (the Out-of-memory just would occur with other important processes)…
3GB RAM seems to be sufficient now; litecoind deamon process is running stable now…I don’t know much about command line nodes.
OK, but thanks to this post I learned how to extract some info about the state of the node (there is no colorful, animated, fancy progress bar in the shell )
I got incoming port 9333 open, with proven connectivity, but no active connections…
So, just to understand the current status of my node: Given the block-time of Litecoin being “2778598” and this equals headers number of ./litecoin-cli --getinfo, this means the node is online, right?
and if blocks number is below headers number means that my node is still syncing with the network or validating old block or something, right? I got a high CPU-load which would also point in this direction…
