Fix: MISCONF Redis is configured to save RDB snapshots, but it’s currently unable to persist to disk

Write commands such as SET or HSET fail with:

(error) MISCONF Redis is configured to save RDB snapshots, but it's currently unable to persist to disk. Commands that may modify the data set are disabled, because this instance is configured to report errors during writes if RDB snapshotting fails (stop-writes-on-bgsave-error option). Please check the Redis logs for details about the RDB error.

Reads still work. With RDB save points configured, the default stop-writes-on-bgsave-error yes makes Redis refuse writes after a background save fails, so you notice that data is no longer reaching disk. The error clears once a snapshot succeeds again.

Step 1: read the real reason

redis-cli INFO persistence | grep -E 'rdb_last_bgsave_status|rdb_changes_since_last_save|rdb_last_bgsave_time_sec'
redis-cli CONFIG GET dir
redis-cli CONFIG GET dbfilename
sudo journalctl -u redis-server -n 100 --no-pager    # or the logfile set in redis.conf

rdb_last_bgsave_status:err confirms the failed save. The log line written at the time of the failure names the cause. The service may be called redis-server, redis or valkey depending on the package; CONFIG GET logfile shows the log path if it is not the journal.

Step 2: fix the cause (most likely first)

1. Disk full or quota reached. Check the filesystem that holds dir with df -h /var/lib/redis (use your dir). A snapshot needs room for a complete new file next to the old one before the rename. Free space or move dir to a larger volume.
2. Permissions. The Redis user must be able to create files in dir: sudo -u redis touch /var/lib/redis/.write-test && sudo rm /var/lib/redis/.write-test. Fix ownership with chown redis:redis on the directory. Check that dir was not changed at runtime with CONFIG SET dir to a path Redis cannot write; that is also a sign of an exposed, compromised instance.
3. The fork failed or the child was killed. RDB saving forks a child process. Under memory pressure the fork can fail (the log mentions fork or “Cannot allocate memory”), or the kernel OOM killer can end the child (dmesg | grep -i oom). Give the host more free memory, set maxmemory below the available RAM, and follow the overcommit advice Redis prints in its log at startup.
4. Read-only or failing disk. Kernel messages in dmesg about I/O errors or a filesystem remounted read-only mean the storage itself needs attention first.

Step 3: prove it works

redis-cli BGSAVE
sleep 5
redis-cli INFO persistence | grep rdb_last_bgsave_status    # want: ok
redis-cli SET misconf:test ok EX 60

Once rdb_last_bgsave_status is ok, writes are accepted again without a restart.

The emergency switch, and why to be careful

Redis’s own knowledge base describes a workaround that lets writes through while the disk problem persists:

redis-cli CONFIG SET stop-writes-on-bgsave-error no

This does not fix anything: Redis keeps accepting writes it cannot save, and a restart or crash loses them. Use it only to keep a pure cache running, or for a short time while you fix the disk, and only with monitoring on rdb_last_bgsave_status. Set it back to yes afterwards. If the instance is only a cache and you never need its data after a restart, the cleaner choice is to disable snapshots entirely with save "".

AOF variant

With AOF enabled, a failing write to the append-only file produces a related error that starts with MISCONF Errors writing to the AOF file: followed by the system error. The causes and checks are the same: disk space, permissions and disk health, plus aof_last_write_status in INFO persistence.

Official documentation

• Redis knowledge base: How to fix the MISCONF error
• Redis docs: Persistence