baiglin opened a new issue, #644:
URL: https://github.com/apache/mina-sshd/issues/644
### Version
2.13.2
### Bug description
Hi @tomaswolf,
I have a strange situation where I get this exception:
org.apache.sshd.common.io.WritePendingException: A write operation is
already pending; cannot write ... bytes.
To explain the context, I use a shared session:
```
ClientSession clientSession =
this.sshClient
.connect(sshdConfiguration.getUserName(),
sshdConfiguration.getHost(),
sshdConfiguration.getPort())
.verify(verifyTimeout)
.getSession();
```
And then in order to not create too many sessions I create different
channels on this same session:
```
SftpClient client =
SftpClientFactory.instance().createSftpClient(clientSession);
```
The different clients are used in one thread only, there is no concurrency
involved when using them… So for me I should not have parallels write
operations on the same client so not on the same `ChannelAsyncOutputStream` …
Still I get this exception… What I can see though is it seems, I first get a:
```
Caused by: org.apache.sshd.common.SshException:
IoWriteFutureImpl[SftpChannelSubsystem[id=8,
recipient=8]-ClientSessionImpl[zhongyuanftp@/8.149.140.159:22][sftp][SSH_MSG_CHANNEL_DATA]]:
Failed to get operation result within specified timeout: 30000 msec
at
org.apache.sshd.common.future.AbstractSshFuture.lambda$verifyResult$1(AbstractSshFuture.java:114)
~[sshd-common-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.common.future.AbstractSshFuture.formatExceptionMessage(AbstractSshFuture.java:206)
~[sshd-common-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.common.future.AbstractSshFuture.verifyResult(AbstractSshFuture.java:114)
~[sshd-common-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.common.io.AbstractIoWriteFuture.verify(AbstractIoWriteFuture.java:41)
~[sshd-common-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.common.io.AbstractIoWriteFuture.verify(AbstractIoWriteFuture.java:32)
~[sshd-common-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.common.future.VerifiableFuture.verify(VerifiableFuture.java:110)
~[sshd-common-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.common.future.VerifiableFuture.verify(VerifiableFuture.java:96)
~[sshd-common-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.sftp.client.SftpMessage.waitUntilSent(SftpMessage.java:85)
~[sshd-sftp-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.sftp.client.impl.SftpOutputStreamAsync.internalFlush(SftpOutputStreamAsync.java:358)
~[sshd-sftp-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.sftp.client.impl.SftpOutputStreamAsync.internalTransfer(SftpOutputStreamAsync.java:285)
~[sshd-sftp-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.sftp.client.impl.SftpOutputStreamAsync.write(SftpOutputStreamAsync.java:180)
~[sshd-sftp-2.14.0.jar!/:2.14.0]
```
Then I get when closing the channel or trying to send another message later:
```
org.apache.sshd.common.io.WritePendingException: A write operation is
already pending; cannot write 31 bytes
at
org.apache.sshd.common.channel.ChannelAsyncOutputStream.writeBuffer(ChannelAsyncOutputStream.java:110)
~[sshd-core-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.sftp.client.impl.DefaultSftpClient.write(DefaultSftpClient.java:308)
~[sshd-sftp-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.sftp.client.impl.DefaultSftpClient.send(DefaultSftpClient.java:272)
~[sshd-sftp-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.sftp.client.impl.AbstractSftpClient.rpc(AbstractSftpClient.java:167)
~[sshd-sftp-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.sftp.client.impl.AbstractSftpClient.checkCommandStatus(AbstractSftpClient.java:232)
~[sshd-sftp-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.sftp.client.impl.AbstractSftpClient.close(AbstractSftpClient.java:609)
~[sshd-sftp-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.sftp.client.impl.DefaultCloseableHandle.close(DefaultCloseableHandle.java:53)
~[sshd-sftp-2.14.0.jar!/:2.14.0]
at
org.apache.sshd.sftp.client.impl.SftpOutputStreamAsync.close(SftpOutputStreamAsync.java:440)
~[sshd-sftp-2.14.0.jar!/:2.14.0]
```
I wonder if the internal state
`ChannelAsyncOutputStream#writeState.writeInProgress = true;` is not reset
leading to that problem ?
I can add that I check before trying a new transfer if the channel is
closed, so if my first guess is correct I guess we should have the channel
closed in such case or the state reset ?
Also if you could confirm that the first timeout is caused by the fact that
server is not replying at all or at least we did not receive a reply ?
### Actual behavior
We get a time out then exception on next write
### Expected behavior
I expected to be able to write again on the same channel or get the channel
closed
### Relevant log output
_No response_
### Other information
_No response_
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
