Status Report for SLAC/DOE IEPM Project

For September & October 1997

SLAC Home Page


Usual assistance for new employee (Warren Matthews) starting on new project, bringing him up to speed, getting squared away with personnel, getting environment set up etc.


Prepared presentation on IEPM/ICFA monitoring activities for the ESCC meeting at JLAB at beginning of October. Warren & Les attended the meeting and Les presented the talk. At the meeting we had many useful discussions related to the IEPM work, with:

With assistance from others came up with a "Brand Name" for our Ping tools. The suite of tools are now known as PingER and have an icon (a dolphin with sound waves being emitted from its head, see

Prepared a Web page providing information on how to get and install the PingER tools, and sent the information to the 6 XIWT/IPWT sites who want to install Pinger. Worked with Jeff Sedayao of Intel to successfully port some of the tools to BSDI. Made sure the changes got back into the master source. Three of the 6 sites have the collection tools installed, and the remainder hopes to be done by early November. They have more concerns about security than we do, so that will be a challenge. Les & Warren attended a voice conference of the XIWT/IPWT. The main discussion was on the deployment/porting of Pinger to their sites, plus the agenda for the face to face meeting in Santa Clara on November 13th.

Worked with Guy Almes & Vern Paxson to set some guidelines for installation of a Surveyor at SLAC.

Worked with DESY to encourage them and get them up and running as a PingER collection site.

After hearing that our Web page on Traceroute Servers for HENP & ESnet (see had made Yahoo's Top Ten list for Internet and Computers, we updated the page to ensure it is current.

With HEPNRC we finished preparing and reformating a paper on Esnet WAN monitoring. This is for publication in an Elsevier book on the CHEP97 conference in Berlin.


Worked with Franca Fiumana of INFN/CNAF to improve the ping_data_plot script to do a better job of cleaning up temporary files and to avoid trying to plot too much data.

The Connectivity script was enhanced to add a slope (in kbytes / sec) based on the ping response times for 100 and 1000 byte pings.

Extended pinghistory to allow for SAS tables that wrap lines into a second page due to line length limitations. Alos finished off making pinghistory work with the Unreachability, Unpredictability, and Quiescent Network SAS tables. Started work on extending pinghistory's configurability to enable it to handle the 10 week alert table. (see:

We added a function to timeping to calculate the median ping response time, including special consideration for pathological replies. This has been fed back to HEPNRC for official incorporation and release.

More work was done by Lois White, with guidance from Les, to modify the SAS long term history table generation scripts to enable them to work with the new data format being collected from multiple collection sites.

Back to Top

Revised 23 October 1997
Comments to