You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mina.apache.org by "Goldstein Lyor (JIRA)" <ji...@apache.org> on 2018/02/01 18:02:00 UTC
[jira] [Resolved] (SSHD-800) Numerous SSH_MSG_CHANNEL_FAILURE
messages sent for SSH_MSG_CHANNEL_DATA on unknown channel
[ https://issues.apache.org/jira/browse/SSHD-800?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Goldstein Lyor resolved SSHD-800.
---------------------------------
Resolution: Fixed
Assignee: Goldstein Lyor
Fix Version/s: 1.8.0
Made the value configurable - and after some consideration decided to make the default _false_ - i.e., don't reply to such messages - which means you don't need special SSHD configuration
> Numerous SSH_MSG_CHANNEL_FAILURE messages sent for SSH_MSG_CHANNEL_DATA on unknown channel
> -------------------------------------------------------------------------------------------
>
> Key: SSHD-800
> URL: https://issues.apache.org/jira/browse/SSHD-800
> Project: MINA SSHD
> Issue Type: Bug
> Affects Versions: 1.7.0
> Reporter: Fulvio Cavarretta
> Assignee: Goldstein Lyor
> Priority: Major
> Fix For: 1.8.0
>
>
> Hi in case the SSHD server receives a message for a not existing channel, it sends back a _SSH_MSG_CHANNEL_FAILURE_.
> This behaviour is not described in RFC4254, which, I agree, is unclear about what should be do in this case.
> Since the current SSHD implementation send back a _SSH_MSG_CHANNEL_FAILURE_ message (with the local channel id), in our environment, it put the receipient in an inconsistent state.
> In out enviroment happened that the receipient has sent a _SSH_MSG_CHANNEL_DATA_ while SSHD had already closed the channel.
> I fully agree that this should not happen (I will open another issue about this) but the _SSH_MSG_CHANNEL_FAILURE_ further complicates things.
>
> The problem can be overcome modifying line 78 of _org.apache.sshd.common.session.helpers.DefaultUnknownChannelReferenceHandler_
>
> letting
> {code:java}
> case SshConstants.SSH_MSG_CHANNEL_EXTENDED_DATA:
> // Not sure if entirely compliant with RFC4254, but try to stem the flood
> wantReply = false;
> break;
> default: // do nothing{code}
>
> instead of
>
> {code:java}
> case SshConstants.SSH_MSG_CHANNEL_EXTENDED_DATA:
> // Not sure if entirely compliant with RFC4254, but try to stem the flood
> wantReply = true;
> break;
> default: // do nothing{code}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)