SIGSEGV received, aborting Aerospike Community Edition build 4.2.0.3 os el6

[Sachin]

Issue: I am seeing following error whenever I scan my secondary index (geo2dsphere). Can someone point me to the possible reason?

Configuration File:

/var/log/aerospike.log

System Information: NAME=“Amazon Linux AMI” VERSION=“2018.03” ID=“amzn” ID_LIKE=“rhel fedora” VERSION_ID=“2018.03” PRETTY_NAME=“Amazon Linux AMI 2018.03” ANSI_COLOR=“0;33” CPE_NAME=“cpe:/o:amazon:linux:2018.03:ga” HOME_URL="AWS | Amazon Linux AMI

Aerospike Information: Aerospike Community Edition build 4.2.0.3

[pgupta]

1,055,503 exceeds allowable max - 1,048,576 (1 MB) [ Known Limitations | Aerospike Documentation ]

[kporter]

Even so, it should not cause a SEGV.

Aerospike 4.2.0.3 is beyond our support window. I recommend upgrading - the sindex-related modules have received a lot of attention in recent development cycles, I know there were a couple of failure path issues addressed in recent releases.

[Sachin]

Thanks for reply.

Yes, I put that validation now while querying in Aerospike. Still the aerospike is aborting with the message “SIGSEGV received, aborting Aerospike Community Edition build 4.2.0.3 os el6”.

Please help, thanks.

[Sachin]

Thanks for reply.

I have upgraded to Aerospike Community Edition build 4.6.0.4 but still the issue is coming.

I am running a job which does around 100 transactions per second and the activity seems fine for 2 hours and after that the Aerospike suddenly abort.

[kporter]

I’d suggest upgrading to the latest 6.0 or possibly 5.7 versions of the server. The secondary index and query code paths underwent significant changes in 6.0 and 5.7.

BTW, it is odd that you upgraded to 4.6.0.4 instead of taking the latest hotfix release in the 4.6.0.x lineage which is currently 4.6.0.21. Moving to the latest hotfix release for a particular lineage is practically always safe. Hotfixes do not introduce new features and are typically very minimal changes to address known issues.