Closed snowyaaron1 closed 2 years ago
So today I decided to comment out wait_for_nodetool_up() and commented out Line 966, rebuilt the image as a local copy and it worked, I am able to cluster between the two using this new image
IP: 10.117.50.47
services:
cassandra:
image: test
ports:
- 7000:7000
- 7001:7001
- 9042:9042
environment:
#- CASSANDRA_HOST=cassandra
- CASSANDRA_SEEDS=10.117.50.41,10.117.50.47
- CASSANDRA_CLUSTER_NAME=cassandra-cluster
- CASSANDRA_PASSWORD=cassandra
- CASSANDRA_BROADCAST_ADDRESS=10.117.50.47
- BITNAMI_DEBUG=true
# By default, Cassandra autodetects the available host memory and takes as much as it can.
# Therefore, memory options are mandatory if multiple Cassandras are launched in the same node.
- MAX_HEAP_SIZE=512M
- HEAP_NEWSIZE=256M
Hi @snowyaaron1, I'm not sure that disabling the wait_for_nodetool_up
function is a good idea. Have you considered entering into one of those pods and executing the nodetool status
command for yourself? That might reveal your issue.
You can also change the following line in libcassandra.sh
to reveal what it is checking against:
@@ -840,7 +840,9 @@ wait_for_nodetool_up() {
if [[ "$BITNAMI_DEBUG" = "true" ]]; then
output="/dev/stdout"
fi
+ set -x
"${check_cmd[@]}" "${check_args[@]}" | grep -E "${check_regex}" >"${output}"
+ set +x
}
if retry_while check_function_nodetool "$retries" "$sleep_time"; then
After running nodetool status
it shows that neither are connecting to each other. The Bitnami images continues to fail (exits container) on wait_for_nodetool_up
with no error message. I'm curious why if I switch the image to the Apache Cassandra one with the exact same docker-compose above it works perfectly (they connect to one another). Or, if I take out wait_for_nodetool_up
it also works.
Hi, Have you tried to enable BITNAMI_DEBUG ? or running the check command manually ? Maybe that gives a clue of what is happening.
Bitnami crashes on this error:
cassandra 20:52:18.01 ERROR ==> Nodetool output cassandra 20:52:18.00 ERROR ==> Cassandra failed to start up
Hi, Have you tried creating a network and using the same network in both docker-compose ? My guess is that this issue is more related to a misconfiguration of cassandra/docker than a issue in the image itself.
Well i thought the same thing, These are two separate instances over AWS. If i change the image to Apache Cassandra it works perfectly. It's something bitnami is doing
We are going to transfer this issue to bitnami/containers
In order to unify the approaches followed in Bitnami containers and Bitnami charts, we are moving some issues in bitnami/bitnami-docker-<container>
repositories to bitnami/containers
.
Please follow bitnami/containers to keep you updated about the latest bitnami images.
More information here: https://blog.bitnami.com/2022/07/new-source-of-truth-bitnami-containers.html
Sorry for the delay. Could you check the listening IP in the image that works for you ? I guess both images generates a very similar configuration.
This Issue has been automatically marked as "stale" because it has not had recent activity (for 15 days). It will be closed if no further activity occurs. Thanks for the feedback.
Due to the lack of activity in the last 5 days since it was marked as "stale", we proceed to close this Issue. Do not hesitate to reopen it later if necessary.
Name and Version
docker.io/bitnami/cassandra:4.0.4
What steps will reproduce the bug?
Run using docker-compose on 2 separate instances. IP: 10.117.50.41
IP: 10.117.50.47
What is the expected behavior?
Connect between two nodes via handshake and gossip
If I switch the image to "cassandra:4.0.4" the default Cassandra image this example gossips accordingly
What do you see instead?
Bitnami crashes on this error:
cassandra 20:52:18.01 ERROR ==> Nodetool output cassandra 20:52:18.00 ERROR ==> Cassandra failed to start up
BUT ITS EMPTY UNDER HERE
Additional information
I am not creating any docker networks. Im allowing docker to assign the IP at docker-compose up