mina-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Goldstein Lyor (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (SSHD-783) mina sshd Nio2Acceptor seems like memory peak
Date Mon, 06 Nov 2017 16:56:00 GMT

    [ https://issues.apache.org/jira/browse/SSHD-783?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16240552#comment-16240552

Goldstein Lyor commented on SSHD-783:

We no longer issue patches for version 1.0 - please try version 1.6 (latest) and see if the
issue persists

> mina sshd Nio2Acceptor seems like memory peak
> ---------------------------------------------
>                 Key: SSHD-783
>                 URL: https://issues.apache.org/jira/browse/SSHD-783
>             Project: MINA SSHD
>          Issue Type: Question
>    Affects Versions: 1.1.0
>         Environment: Linux version 3.10.0
> jdk-1.7.0_76
>            Reporter: young_shaw
>            Priority: Critical
>              Labels: sshd
>         Attachments: 1.png
>   Original Estimate: 48h
>  Remaining Estimate: 48h
> I am using sshd 1.1.0 as an sftp server running on a Red Hat os. I found this program
often produces hprof file. I use MAT to analysis this file and find this kind of prompt:
> "org.apache.sshd.common.io.nio2.Nio2Acceptor" instance "0x761551a68" occupies 3,577,793,864
(98.18%) byte. Its main accumulation of memory is "java.util.concurrent.ConcurrentHashMap$Segment[]"
instance "0x7608721c8" loaded by "bootstrap class loader" .
> Futher I found Nio2Accepter's sessions is so huge, seen in the attachment1. I doubt the
method "sessionClosed" in Nio2Service is not executed when a client disconnect abnormally.
So I want to ask your help to postion this problem. Thank you.

This message was sent by Atlassian JIRA

View raw message