docs/vmbackup.md: mention that -filestream.disableFadvise command-line flag can be used for reducing CPU usage on systems with big number of CPU cores

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/pull/5120

This is a follow-up for 75dd7b30ba
This commit is contained in:
Aliaksandr Valialkin 2023-10-04 16:36:40 +02:00
parent d863d11b5a
commit ade64b5783
No known key found for this signature in database
GPG Key ID: A72BEC6CD3D0DED1
2 changed files with 4 additions and 2 deletions

View File

@ -136,7 +136,8 @@ See [this article](https://medium.com/@valyala/speeding-up-backups-for-big-time-
## Troubleshooting ## Troubleshooting
* If the backup is slow, then try setting higher value for `-concurrency` flag. This will increase the number of concurrent workers that upload data to backup storage. * If the backup is slow, then try setting higher value for `-concurrency` flag. This will increase the number of concurrent workers that upload data to backup storage.
* If `vmbackup` eats all the network bandwidth, then set `-maxBytesPerSecond` to the desired value. * If `vmbackup` eats all the network bandwidth or CPU, then either decrease the `-concurrency` command-line flag value or set `-maxBytesPerSecond` command-line flag value to lower value.
* If `vmbackup` consumes all the CPU on systems with big number of CPU cores, then try running it with `-filestream.disableFadvise` command-line flag.
* If `vmbackup` has been interrupted due to temporary error, then just restart it with the same args. It will resume the backup process. * If `vmbackup` has been interrupted due to temporary error, then just restart it with the same args. It will resume the backup process.
* Backups created from [single-node VictoriaMetrics](https://docs.victoriametrics.com/Single-server-VictoriaMetrics.html) cannot be restored * Backups created from [single-node VictoriaMetrics](https://docs.victoriametrics.com/Single-server-VictoriaMetrics.html) cannot be restored
at [cluster VictoriaMetrics](https://docs.victoriametrics.com/Cluster-VictoriaMetrics.html) and vice versa. at [cluster VictoriaMetrics](https://docs.victoriametrics.com/Cluster-VictoriaMetrics.html) and vice versa.

View File

@ -147,7 +147,8 @@ See [this article](https://medium.com/@valyala/speeding-up-backups-for-big-time-
## Troubleshooting ## Troubleshooting
* If the backup is slow, then try setting higher value for `-concurrency` flag. This will increase the number of concurrent workers that upload data to backup storage. * If the backup is slow, then try setting higher value for `-concurrency` flag. This will increase the number of concurrent workers that upload data to backup storage.
* If `vmbackup` eats all the network bandwidth, then set `-maxBytesPerSecond` to the desired value. * If `vmbackup` eats all the network bandwidth or CPU, then either decrease the `-concurrency` command-line flag value or set `-maxBytesPerSecond` command-line flag value to lower value.
* If `vmbackup` consumes all the CPU on systems with big number of CPU cores, then try running it with `-filestream.disableFadvise` command-line flag.
* If `vmbackup` has been interrupted due to temporary error, then just restart it with the same args. It will resume the backup process. * If `vmbackup` has been interrupted due to temporary error, then just restart it with the same args. It will resume the backup process.
* Backups created from [single-node VictoriaMetrics](https://docs.victoriametrics.com/Single-server-VictoriaMetrics.html) cannot be restored * Backups created from [single-node VictoriaMetrics](https://docs.victoriametrics.com/Single-server-VictoriaMetrics.html) cannot be restored
at [cluster VictoriaMetrics](https://docs.victoriametrics.com/Cluster-VictoriaMetrics.html) and vice versa. at [cluster VictoriaMetrics](https://docs.victoriametrics.com/Cluster-VictoriaMetrics.html) and vice versa.