Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [milo-dev] ChunkDecoder#decode() error message 
The content of the log is that the sequence number is always
different by one, but it is hard to keep track of including
the Netty layer.

Someday I'd like to make time and do a heat-run test with 0.2.x
server and client. I am glad if someone could do the heat-run test
for at least a week. Or until this kind of error occurs.

After that, I will try 0.3.x that will be changed to Netty 4.1.x ?
By the way I am very expected for 0.3.x.

On 2018/02/14 22:21, Kevin Herron wrote:

The problem is even if it occurs after 0.2.0 I have no way of knowing if
it’s a bug or if it’s correctly catching a missing chunk.

On Tue, Feb 13, 2018 at 19:28 Shigeru Ishida <ishida_shigeru@xxxxxxxxxxx>
wrote:


If it does not occur after 0.2.0, we will consider in this direction in
the future.

Thank you for the information.

On 2018/02/14 12:03, Kevin Herron wrote:

I've seen that error before and never figured out if it was actually a

bug

in Milo (I've looked many times) or the sequence numbers on each chunk

and

the required check for continuity doing its job: i.e. a chunk was lost

due

to a dropped packet or some other error scenario in the vast amount of
networking abstractions sitting below Milo.

Starting in 0.2.0 the chunk handling was refactored - it might be worth
seeing if the same error happens there as well.

On Tue, Feb 13, 2018 at 6:04 PM, Shigeru Ishida <

ishida_shigeru@xxxxxxxxxxx>

wrote:


Hi Kevin,

I may have asked a similar question before. Sometimes

ChunkDecoder#decode()

error message is output when running the server.

These servers are developed using Milo 0.1.6, and connected clients are
also
created using 0.1.6. To these servers, 5-6 clients are connected at the
same time
and reading data.

1) Lightweight server without Apache Felix
    Continuous running exceeded 35 days, and 13 error logs have been

output

so far.

2) Lightweight server with Apache Felix 5.6.10 (latest)
    Continuous running exceeded 22 days and error log was output three

times

so far.

In addition, 1) servers have more clients connected than 2). 1), in
reality,
more than 6 clients, including clients from other companies, are

connected.

In addition, data acquisition from clients of other companies is not
frequent.

2), only clients created in Milo 0.1.6 are connected.

Sample error logs are as follows. The error log that is output is of

this

type only.
---
2018-02-12 00:03:03,571+JST 重大

org.eclipse.milo.opcua.stack.core.channel.ChunkDecoder

decode - expected sequence number 3693718 but received 3693719
2018-02-12 00:03:03,573+JST 重大

org.eclipse.milo.opcua.stack.core.channel.ChunkDecoder

decode - 4d5347467200000008000000e0250000975c3800a388db0001003a030500
0020000000e5bdbdadf74e1d7c2ba375bd155b4c5cf3b6939ff1478ebdb4
1954cef6f4acee00c49b8349a3d3016e84db0000000000ffffffffa86100
00000000020000001900000006fa51001900000007fa5100
2018-02-12 00:03:03,574+JST 重大 org.eclipse.milo.opcua.stack.s
erver.handlers.UaTcpServerSymmetricHandler lambda$1 - Error decoding
symmetric message: expected sequence number 3693718 but received 3693719
UaException: status=Bad_SecurityChecksFailed, message=expected sequence
number 3693718 but received 3693719
          at org.eclipse.milo.opcua.stack.core.channel.ChunkDecoder.decod
e(ChunkDecoder.java:137)
          at org.eclipse.milo.opcua.stack.core.channel.ChunkDecoder.decod
eSymmetric(ChunkDecoder.java:80)
          at org.eclipse.milo.opcua.stack.server.handlers.UaTcpServerSymm
etricHandler.lambda$1(UaTcpServerSymmetricHandler.java:177)
          at org.eclipse.milo.opcua.stack.core.channel.SerializationQueue
.lambda$1(SerializationQueue.java:58)
          at org.eclipse.milo.opcua.stack.core.util.ExecutionQueue$PollAn
dExecute.run(ExecutionQueue.java:107)
          at java.util.concurrent.Executors$RunnableAdapter.call(
Executors.java:511)
          at java.util.concurrent.FutureTask.run(FutureTask.java:266)
          at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPool
Executor.java:1142)
          at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoo
lExecutor.java:617)
          at java.lang.Thread.run(Thread.java:748)
---

Although the server is running continuously, it is worrisome because the
log level
is ERROR.

Previously, I was explained that this type of error log may be related

to

memory leak.

I do not think there is urgency, but if there is a hint of how to deal
with something,
I am glad that you tell me helpful information. I will refer to it and
consider
future countermeasures.

Regards,

--Shigeru
_______________________________________________
milo-dev mailing list
milo-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe
from this list, visit
https://dev.eclipse.org/mailman/listinfo/milo-dev





_______________________________________________
milo-dev mailing list
milo-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe

from this list, visit

https://dev.eclipse.org/mailman/listinfo/milo-dev


_______________________________________________
milo-dev mailing list
milo-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe
from this list, visit
https://dev.eclipse.org/mailman/listinfo/milo-dev





_______________________________________________
milo-dev mailing list
milo-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/milo-dev

Back to the top