In 12/99 we quoted and resold PSINet InterMAN T1 service ("full" 1.17 speed) to a customer, but during installation BellSouth absolutely could not provision as prescribed. After several attempts and delays, PSINet offered us 1.54 T1 Frame Conn @ 50% CIR instead. Its unclear now if our ex-sales rep ever communicated change of service to customer, as all parties have departed. New person at customer site now complains of slow speed after 1 year+ of service. Says he's owned dedicated T1 and FR before and this is no T1 speed. Talking about year's worth of "fraud" on our part. Yikes!! PSINet's tests show no hint of any speed problems. Our attempts to test onsite via internet speed tests (timed file download) produce results in the 30% to 50% of expected "T1 speed". How best to monitor slowness, perhaps just to certain sites? Or at certain times? Additional problem in that our cost for FR was actually about $100 higher than cost for InterMAN. I would have thought FR should be lower, but PSINet called it an upgrade (meaning: more $$$ to PSINet). Just looking for comments, suggestions or sympathy.
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations gkittelson on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
Help with unsatisfied Frame User 1
- Thread starter tneidepxe
- Start date
Get a sniffer to track the actual packets coming into the router and coming back from the router. Use the counters on the routers for basic numbers, Something like "monitor" from solarwinds.net and provide realtime SNMP based traffic monitoring( pretty accurate). Use the packet generator or "WAN-Killer" from Solarwinds to load the link and do the math when it maxes out. PSInet *should* have something that can spit up number of utilization of the frame circuit. Have the user run their "slow" app and get a sniff trace and look at the relative and absolute times to see if the delay is local( server clog) or WAN.
None is hard,.. just a bit labor intensive :/
Mike S
"Diplomacy; the art of saying 'nice doggie' till you can find a rock" Wynn Catlin
None is hard,.. just a bit labor intensive :/
Mike S
"Diplomacy; the art of saying 'nice doggie' till you can find a rock" Wynn Catlin
-
1
- Thread starter
- #3
Thanks Wyb - we'll try that. Internally they're on 100Mb fiber with only a couple of legs at 10Mb; where we had to use existing copper and Tutsystems. All his reports of slowness are with Internet access/download times - not with apps - seemingly no matter where within the network. I was about to ask how reliable the myriad of Internet Speed Tests out there are.. but I just did a quick test. I'm sitting here at home on a modem with a solid 50.6k connection and it says I test at 670-679k; nearly half of the "T1" mark. That's about the same results I get from the customer's site, or from our own T1 at the office (non PSINet). So to answer my own question - those tests are pretty much useless or at best misleading.
PSINet says likelihood of any partial refund for past year's time, is "slim" - mainly because no speed related problems were ever reported, only 2 or 3 outages due to Bell and they were fixed in reasonable time.
Question - Were we rubes for letting PSINet & Bell off the hook for the original circuit and taking the FC alternate? Looks like our ex-salesman may have sold us a bill of goods just as well as the customer. Thanks again.
PSINet says likelihood of any partial refund for past year's time, is "slim" - mainly because no speed related problems were ever reported, only 2 or 3 outages due to Bell and they were fixed in reasonable time.
Question - Were we rubes for letting PSINet & Bell off the hook for the original circuit and taking the FC alternate? Looks like our ex-salesman may have sold us a bill of goods just as well as the customer. Thanks again.
I get the feeling you are left holding the bag. PSI should not have ever been let off the hook if they could not meet the SLA they agreed to. One thing to ask PSI is if you can "burst" over the CIR. Not all providers will let you do that and some tack on charges if you do. Does the interfaces should bits eligable for discard? I doubt the customer could prove "fraud".. since he/she has been using for a year. I would get with a lawyer and go over the contact to see what your liablities really are and not what you think they are. Also find out your legal standing regarding PSInet and the SLA. As far as the law is concerned, the fact people leave the company doesnt let them off the hook.
Good luck!!
Mike S
"Diplomacy; the art of saying 'nice doggie' till you can find a rock" Wynn Catlin
Good luck!!
Mike S
"Diplomacy; the art of saying 'nice doggie' till you can find a rock" Wynn Catlin
- Thread starter
- #5
I think we've hashed out the worst of it on a conference call today. After only a brief spate of raised voices and being called a "liar" only once.... (not bad considering the customer is a Baptist church and we were talking to a pastor) we seem now to be working toward a solution where they will record what sites they visit and what they perceive as their "speed". We'd love nothing better than to remove ourselves from between PSINet and the customer - and that looks like a possibility or hooking him up with another provider which will probably be cheaper anyway. We will also do the things with the sniffer you suggested to prove our side. Some of his own people claim to have downloaded files at 130KBps - roughly 1.3Mbps - right? But he just pooh-poohs that...claims the fellow misspoke himself and that everyone around him are "imbeciles" regarding the Internet - too stupid to realize how slow they were until he showed up. He claims he doesn't ever get above 28.8 speed - his "modem from home is faster"..... I don't think it will come to lawyers -- we'll just get him off on another provider and let them beat each other up directly. Again, thanks Mike for the help and commiseration...
When someone starts saying he/she is the *expert* and everyone around them are the idiots.. I get very worried ;-)
I hope it works out well for you. Been there.. done that.. wont do it again !!
And yes,.. different sites will be different speeds. If you have a T3 and the other site has a P166 on a 128K link.. you're kinda of limited.
Mike S
"Diplomacy; the art of saying 'nice doggie' till you can find a rock" Wynn Catlin
I hope it works out well for you. Been there.. done that.. wont do it again !!
And yes,.. different sites will be different speeds. If you have a T3 and the other site has a P166 on a 128K link.. you're kinda of limited.
Mike S
"Diplomacy; the art of saying 'nice doggie' till you can find a rock" Wynn Catlin
I am having exactly the same problems with Bellsouth/Qwest.
What finally happened ? My line tests clean but when I check the bandwidth with qcheck I get sometimes as low as 50k and I have a 1.54 with 128 CIR. The weird thing is if I test from the remote to the host the throughput is good, just from the host to the remote that I am having problems.
What finally happened ? My line tests clean but when I check the bandwidth with qcheck I get sometimes as low as 50k and I have a 1.54 with 128 CIR. The weird thing is if I test from the remote to the host the throughput is good, just from the host to the remote that I am having problems.
A pretty good site that has some great tools to test internet speed is
It seems pretty accurate.
It seems pretty accurate.
- Status
Similar threads
- Locked
- Question
- Locked
- Question
- Locked
- Question