Hi, We have a problem Configuration is the same as I posted before in thread Aerospike version now is 3.3.17 no anomalies in logs Aerospike server responds to the client “java.lang.Throwable: com.aerospike.client.AerospikeException: Error Code 8: Server memory error” Console tells us that we ar…

Hi, After looking at your configuration on the other thread, I can’t find any settings for your defrag-lwm-pct. Can I have you post the actual configuration from Aerospike server and the last 2000 logs? For Aerospike log: tail -2000 /var/log/aerospike/aerospike.log It seems that Aerospike migh…

You are running with extremely unsafe settings of 99% HWM! You should be running with 50-60% HWM. Also, Your avail pct is on 4% which will hit stop-writes true. For practical purpose, avail pct is more critical than free disk pct. Also, how much is your memory consumption. Can you post a screenshot…

Yes default ttl 0, we use our job for cleanup old data.

I add 1 node to cluster. All 5 node on that moment have some configuration allow-inline-transactions true auto-dun false auto-undun false b…

I cant post full log to discuss. It limit 32000. Logs repeated like this Sep 17 2014 11:03:04 GMT: INFO (drv_ssd): (drv_ssd.c::2392) device /dev/sdc4: used 111389286016, contig-free 43048M (344391 wblocks), swb-free 1, w-q 0 w-tot 1694467 (2.8/s), defrag-q 0 defrag-tot 473966 (0.0/s) Sep 17 2014…

On that moment: How do I clean the disks of excess garbage? Garbage occupies almost 20% of the cluster.

Stats: batch_errors 0 batch_initiate 0 batch_queue 0 batch_timeout 0…

Hi, Can I know what your defrag-lwm-pct is set to? You can grab that info out of the aerospike.conf. Or if not you can also run the following command : asmonitor -e “asinfo -v “namespace/ssd”” We need to know what your defrag-lwm-pct is set to. Since your defrag is not keeping up with your writes…

Hi Previously set in “50” We had structural changes, and the entire cluster was restarted. Now is “99” ~ # asmonitor -e “asinfo -v “namespace/ssd”” Enter help for commands 3 hosts in cluster: host1:3000,host2:3000,host3:3000 -v namespace/ssd host1:3000 returned : type=device objects=16190785…

I would recommend moving your defrag-lwm-pct back to 50. By putting defrag-lwm-pct to that high of a percentage (99) means more blocks to be defragmented, and more dense data on the disk which may lead to latency. Also is there a reason why your stop writes, high-water-disk-pct and high-water-memor…

Unused data on the disk - Error Code 8: Server memory error - #12 by kporter - Configuration

Topic		Replies	Views	Activity
Memory errors Operations	6	4009	July 2, 2017
Bulk/Batch Updates (AER-6499) Delivered Requests	19	7768	June 9, 2022
Server Memory Error (Error Code 8) Configuration error	5	9720	October 27, 2015
Questions about very low disk-avail-pct	14	2921	November 12, 2019
Aerospike in-memory DB uses much memory than expected	23	7626	July 11, 2022

Memory errors

Operations

6

4009

July 2, 2017

Bulk/Batch Updates (AER-6499)

Delivered Requests

19

7768

June 9, 2022

Server Memory Error (Error Code 8)

Configuration

error

5

9720

October 27, 2015

Questions about very low disk-avail-pct

14

2921

November 12, 2019

Aerospike in-memory DB uses much memory than expected

23

7626

July 11, 2022

Unused data on the disk - Error Code 8: Server memory error

Related topics