[Bug] NumberFormatException when consuming messages from pulsar-admin CLI

[david-streamlio]

Search before asking

• I searched in the issues and found nothing similar.

Read release policy

• I understand that unsupported versions don't get bug fixes. I will attempt to reproduce the issue on a supported version of Pulsar client and Pulsar broker.

Version

image: apachepulsar/pulsar-all:3.2.2
platform: linux/amd64

Minimal reproduce step

./bin/pulsar-client consume -n 0 -p Earliest -s sub

What did you expect to see?

No errors

What did you see instead?

$ ./bin/pulsar-client consume -n 0 -p Earliest -s sub persistent://feeds/realtime/coinbase-rfq
2024-04-02T18:34:44,234+0000 [main] WARN org.apache.pulsar.common.util.netty.DnsResolverUtil - Cannot get DNS TTL settings
java.lang.NumberFormatException: For input string: "1 RUN apt-get -y --purge autoremove"
at java.lang.NumberFormatException.forInputString(NumberFormatException.java:67) ~[?:?]
at java.lang.Integer.parseInt(Integer.java:668) ~[?:?]
at java.lang.Integer.valueOf(Integer.java:973) ~[?:?]
at java.lang.Integer.decode(Integer.java:1458) ~[?:?]
at java.util.Optional.map(Optional.java:260) ~[?:?]
at org.apache.pulsar.common.util.netty.DnsResolverUtil.(DnsResolverUtil.java:71) ~[org.apache.pulsar-pulsar-common-3.2.2.jar:3.2.2]
at org.apache.pulsar.client.impl.ConnectionPool.createAddressResolver(ConnectionPool.java:164) ~[org.apache.pulsar-pulsar-client-original-3.2.2.jar:3.2.2]
at org.apache.pulsar.client.impl.ConnectionPool.lambda$new$1(ConnectionPool.java:127) ~[org.apache.pulsar-pulsar-client-original-3.2.2.jar:3.2.2]
at java.util.Optional.orElseGet(Optional.java:364) ~[?:?]
at org.apache.pulsar.client.impl.ConnectionPool.(ConnectionPool.java:127) ~[org.apache.pulsar-pulsar-client-original-3.2.2.jar:3.2.2]
at org.apache.pulsar.client.impl.ConnectionPool.(ConnectionPool.java:96) ~[org.apache.pulsar-pulsar-client-original-3.2.2.jar:3.2.2]
at org.apache.pulsar.client.impl.ConnectionPool.(ConnectionPool.java:91) ~[org.apache.pulsar-pulsar-client-original-3.2.2.jar:3.2.2]
at org.apache.pulsar.client.impl.PulsarClientImpl.(PulsarClientImpl.java:199) ~[org.apache.pulsar-pulsar-client-original-3.2.2.jar:3.2.2]
at org.apache.pulsar.client.impl.PulsarClientImpl.(PulsarClientImpl.java:156) ~[org.apache.pulsar-pulsar-client-original-3.2.2.jar:3.2.2]
at org.apache.pulsar.client.impl.ClientBuilderImpl.build(ClientBuilderImpl.java:65) ~[org.apache.pulsar-pulsar-client-original-3.2.2.jar:3.2.2]
at org.apache.pulsar.client.cli.CmdConsume.consume(CmdConsume.java:149) ~[org.apache.pulsar-pulsar-client-tools-3.2.2.jar:3.2.2]
at org.apache.pulsar.client.cli.CmdConsume.run(CmdConsume.java:141) ~[org.apache.pulsar-pulsar-client-tools-3.2.2.jar:3.2.2]
at org.apache.pulsar.client.cli.PulsarClientTool.run(PulsarClientTool.java:246) ~[org.apache.pulsar-pulsar-client-tools-3.2.2.jar:3.2.2]
at org.apache.pulsar.client.cli.PulsarClientTool.main(PulsarClientTool.java:286) ~[org.apache.pulsar-pulsar-client-tools-3.2.2.jar:3.2.2]

Anything else?

No response

Are you willing to submit a PR?

• I'm willing to submit a PR!

[dao-jun]

@nodece

[lhotari]

Fixed by #22400

[lhotari]

The DNS settings problem has existed for a very long time in the Dockerfile, since #17733 in Pulsar 3.0.0, but it only became visible after #22329 changes. Unfortunately, nobody ran extensive tests with the docker images during the release candidate phase of Pulsar 3.2.2 and Pulsar 3.0.4 . We'll have the fix in Pulsar 3.2.3 and Pulsar 3.0.5 .

[lhotari]

Closing this issue since the fix PR has been merged to branch-3.0 and branch-3.2 .

[lhotari]

The impact of the DNS settings being broken in the Dockerfile since Pulsar 3.0.0 is that the positive DNS cache TTL has been 30 seconds. The negative DNS cache TTL has been 0, so there hasn't been any caching at all.

[visortelle]

Unfortunately, nobody ran extensive tests with the docker images during the release candidate phase

Would having nightly images help to catch more issues before an actual release?
#14755