![]() |
andrew rinaldi <andrew.rinaldi <at> mainpine.com> writes: > > > This sort of detailed information is invaluable Rob, thank > you very much ! <snip> <snip> > On 2/3/07, andrew > rinaldi <andrew.rinaldi <at> mainpine.com> > wrote: > Hi > Mark,'Speed' is not the issue here... With Ethernet it's what happens > when youget large packet loss (which equates to lost 'audio' in our > case)... Andwith virtualisation it's the loss of deterministic behaviour > (which degrades low level timing)... In both cases some systems will work > and some won't...And the 'working' systems can almost always be broken by > heavy work load onthe network or system.A good example of what > happens with Ethernet packet loss is to run a VoIP call across a busy > network and listen to the call quality degrade.... In faxthis quickly > equates to a failed call.Have you ever stress tested your setup > ? How many fax's do you send/receivein a typical day > ? In our lab we've been able to break most (maybe all) virtual > servers that we have tested for use with fax servers.Ultimately, I'm > trying to encourage people to keep it simple because in myexperience that > always works out for the best.RegardsANDREW You are right on target, Andrew, I do have a small setup and have not stress tested it. This is also a low-traffic network with mostly reception of faxes, not transmission. I was hoping that this setup could work for the local government agency (who is migrating to a total VM solution) that wants to get rid of a lot of the fax machines in their organization, but continue to route their residual inbound faxes straight through their central archiving and distribution center. The bottom line in our case sounds like it would lie on the reliability and latency of the local network. I think we'll try the idea out and monitor it, and if reliability becomes an issue then the clear fallback will be the use of local serial ports on the dedicated box and having the remote fax lines redirected to the local building. I do still want to note, however, that (so far) the Avocent brand ESP-2MI serial adapters are working great in this low-volume, controlled environment. And they are still developing their drivers - they are not dead products like I have seen with another brand. Mark ____________________ HylaFAX(tm) Users Mailing List _______________________ To subscribe/unsubscribe, click http://lists.hylafax.org/cgi-bin/lsg2.cgi On UNIX: mail -s unsubscribe hylafax-users-request@xxxxxxxxxxx < /dev/null *To learn about commercial HylaFAX(tm) support, mail sales@xxxxxxxxx*