Martin Stone Davis wrote:

Niklas Bergh wrote:

The real question is: how much of your outgoing bw
is used up by QR:s, how much is used by routing requests, and how much
is used up by "real" data.  I wish I knew how to tell this.



Check outputBytes* in the diagnostics. Low is actual data (trailers) and the others mostly are things like DataRequests and QRs and other things


Lemme make sure I understand this.


 From servlet/nodestatus/diagnostics/outputBytesLow/minute:
 >Events Value Change    Mean Total Value        Mean Time Between Events
 >2072   578015          1.3937197026238224E8    29.047779922779924

From servlet/nodestatus/diagnostics/outputBytes/minute:
>Events Value Change Mean Total Value Mean Time Between Events
>386 649802 1.6627010963104272E8 155.36010362694302


That means that in one minute, I sent out 578015 "real" data bytes and 649802 total? If so, my node is doing very well sending data, despite the large number of QR:s!

-Martin


And if I'm right about the above, shouldn't we change the unnecessarily scary messages on "General Information", like:

> It's normal for the node to sometimes reject connections or requests
> for a limited period. If you're seeing rejections continuously the
> node is overloaded or something is wrong (i.e. a bug).

and (in red):

> [QueryRejecting all incoming requests!]

Most users interpret these to mean there really is something wrong with their node.

Instead, let's show:

* total output bytes/sec (regardless of whether limit is in place)
* percent of limit (if limit exists)
* "raw meat" (i.e. trailers) output bytes/sec
* percent of that to total output bytes OR (better)
  percent of that to the limit

Show this last one as a smoothed running average (over maybe 15 minutes?) and warn the user if it drops below some certain % (say, 67%?). Message might read (in red):

"It's normal for a new node to not contribute much to the network. If your node has been running for more than 24 hours, then something may be configured wrongly."

(SIDE NOTE: AFAIK, a common reason for this is due to using Q instead of Y threads and having maximumThreads set too low (which is default). We should therefore change the default.)


If the node is transient, then say (in red):


"Your node is "transient", which means that it may be more difficult for you to retrieve data from the network. To make your node permanent, see [insert link to FAQ or section README file here]."

Also, for non-transient nodes which have no inbound connections (probably due to a firewall problem), we should add this message to the general information page (in red):

"Your node has no incoming connections, which may indicate your Firewall or NAT is not setup properly. For a new node, this is to be expected. If your node has been running for more than 24 hours, then see [insert link to FAQ or section README file here] to correct this problem."

Probably also we'd need another message if the user is getting no connections at all, but that doesn't happen very often.

I think all the above would address the most common problems I hear about from new users.

-Martin


_______________________________________________ Devl mailing list [EMAIL PROTECTED] http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl

Reply via email to