# A write fail warning

**URL:** <https://discuss.aerospike.com/t/a-write-fail-warning/756>\
**Category:** Aerospike Server Benchmarks\
**Created:** [December 16, 2014, 3:31am UTC](https://discuss.aerospike.com/t/a-write-fail-warning/756 "2014-12-16T03:31:34Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![mceiba](https://sea1.discourse-cdn.com/flex019/user_avatar/discuss.aerospike.com/mceiba/32/81_2.png) [@mceiba](https://discuss.aerospike.com/u/mceiba)\
**Post date:** [December 16, 2014, 3:31am UTC](https://discuss.aerospike.com/t/a-write-fail-warning/756/1 "2014-12-16T03:31:34Z")

</div>

My cluster was configured as all data in memory with HDD for persistence. “WARNING (drv\_ssd): (drv\_ssd.c::4260) {state} write fail: queue too deep: q 65, max 64” is found in the log when I were running workload. What’s that mean?

---

<div class="post-metadata">

**Author:** ![chetanvaity](https://sea1.discourse-cdn.com/flex019/user_avatar/discuss.aerospike.com/chetanvaity/32/76_2.png) [@chetanvaity](https://discuss.aerospike.com/u/chetanvaity)\
**Post date:** [December 19, 2014, 12:32am UTC](https://discuss.aerospike.com/t/a-write-fail-warning/756/2 "2014-12-19T00:32:57Z")

</div>

This means the disk writes are taking too long (the Aerospike write queue is full and is not being emptied fast enough). The underlying reason is simply that you have reached the write throughput of your disk.

---

<div class="post-metadata">

**Author:** ![meher](https://sea1.discourse-cdn.com/flex019/user_avatar/discuss.aerospike.com/meher/32/1644_2.png) [@meher](https://discuss.aerospike.com/u/meher)\
**Post date:** [December 20, 2014, 2:24am UTC](https://discuss.aerospike.com/t/a-write-fail-warning/756/3 "2014-12-20T02:24:15Z")

</div>

You will find more details on this here: [device overload troubleshooting](https://discuss.aerospike.com/t/how-to-troubeshoot-fix-device-overload/486).