Careful using FTP as, particularly on fast links, you may find yourself
measuring disk IO instead.
For downstream TCP measurements, wget -O /dev/null http://blah is a
good way of preventing the disk becoming a bottleneck (at least on the
receiving side).
On 13.11.2013 06:53, Bret Clark wrote
S yet. But that is being worked on I'm sure.
> On the other hand - if you really need MPLS shouldn't you be running
> a
> Cisco or a Juniper?
>
> On Jan 8, 2013, at 18:06, Andrew Jones wrote:
>
>> The software does not do everything that mikrotik's routerOS does.
The software does not do everything that mikrotik's routerOS does.
Where is the MPLS support, something that many people use on routerOS?
On 09.01.2013 10:04, Matt Hoppes wrote:
> Please backup your statement. They are based on Yvatta which is very
> stable. There is actually support for them -
Misconfigured MTU values will produce results like this, particularly
noticeable on SSL-enabled websites.
If you can, try reducing the MTU down to say 1300 Bytes and clamp the MSS
to 1280 on a customer router to see if the problem goes away, if it does,
you'll need to work out where the MTU issue i
Based on the information on robtex.com [1], windstream us AT&T as one of
their upstreams. Windstream need to advise all of their upstream providers
of any new prefixes which are to be advertised through their network, so
there may be some truth to what they are saying although two months is a
ridic
On Thu, 26 Jan 2012 00:25:37 -0500, "Tom DeReggi"
wrote:
--SNIP--
> With that said, We've been looking into Juniper lately, I like that
their
> new lines are all based on the same Juno OS, which is Linux. :-)
--SNIP--
Junos is not Linux, but is a set of processes that run on top of a FreeBSD
ker
Sorry, I misinterpreted what you wrote, your calculations are for an
average utilisation of 50%. That sounds plausible (obviously you have more
of an idea about your own traffic profiles than I do!).
On Tue, 22 Feb 2011 13:52:44 +1100, Andrew Jones wrote:
> Those calculations assume 100% l
Those calculations assume 100% link utilisation all the time, which is
likely far from true. Your actual cost/MB will probably be significantly
higher, even at the NOC.
-Jonesy
On Mon, 21 Feb 2011 17:06:35 -0600, Matt wrote:
> Trying to calculate bandwidth costs.
>
> 1Mbps / 8 bits = 125,000 by
rsyslog is a good alternative to syslog-ng, it backs into mysql and works
well with PHPLogCon.
On Fri, 22 Jan 2010 13:25:10 -0500, Jeremy Parr
wrote:
> 2010/1/22 Adam Kennedy :
>> Splunk is the way to go for something like that.
>
> Splunk is very nice, a reasonably decent free product I have u
That is crazy!
Even with conservative numbers:
16000 texts / 31 days in a month/16 waking hours in a day = 32.25 texts
per hour! That's one every two minutes.
Scottie Arnett wrote:
> Keep in mind that probably most of those were in answer to a text or she
> started and was answered back. So i
Chuck Hogg wrote:
> ...
> I'm not trying to bash Jayson's email. I don't see how you can get
> 130Mbit when the port is 100Mbit.
>
The way Ubiquiti justified this in a forum post was that the 130Mbps is
total throughput (up and down). 100Mbps ethernet at full duplex is
100Mbps in each directi
> 5.470Ghz to 5.650Ghz many channels, 30dB EIRP max with DFS and TPC
> enabled, some exemptions for lower power
> 900Mhz is just 916 to 928Mhz allowed with 30dB max
This is not quite correct, 5470-5600Mhz and 5650-5725Mhz are allowed to
be used, EIRP is limited to 1W averaged over the entire tr
Does anyone have an opinion on the mikrotik R5H cards, particularly
compared to the ubiquiti cards?
lakel...@gbcx.net wrote:
> 5/8" Heliax is the largest diameter you can use for 5.8 GHz. LDF4.5-50
>
>
> Sent from my Verizon Wireless BlackBerry
>
> -Original Message-
> From: David Hulseb
13 matches
Mail list logo