fgk@lll-crg.llnl.gov (Frank G Kienast) (06/26/89)
I am curious as to how an exchange with ESS can have so many bugs and quirks in it. I have noticed the following over the past couple years since we got ESS. Each of the things described below has happened to me at least several times. Sometimes I pick up the phone and their is no dialtone. The touch tone pad works, so there is a voltage, but no tone. If I hang up and retry several times, eventually I get a dial tone. Often this is preceeded by a series of half a dozen or so rapidly ascending tones. Sometimes in the middle of a call (both local and long distance), I get mysteriously disconnected. All of a sudden there is a click and then silence, but the voltage is still there. The party on the other end notes the same thing. Occasionally, the system won't let me program in call forwarding. I dial 72# and get a fast busy (reorder). I dial 73# to make sure I'm not already forwarding, and get the reorder here too. I try hanging up, making other calls first, etc. and forwarding still won't work. Then an hour or so later it cures itself. Sometimes when you dial three digits that are not a local exchange, you get a recording right away. Other times, you don't get the recording until you dial seven digits. Also, no distinction is made between a number that is invalid and one that is valid but requires a 1 first. In fact, the recordings are used interchangably, with the "you must first dial a 1 when calling this number" being used for a few weeks for both, then "your call cannot being completed as dialed" being used for several weeks. If you listen to a recording until it times out (twice), you get a reorder signal. A few seconds later, it starts playing other recordings, such as "All circuits are busy now", or even "The call you have made requires a 25 cent deposit"! When you call an AT&T LD operator (by dialing 10288 0#, for example), you hear four DTMF tones followed by a loud "clunk" followed by about ten more DTMF tones (these tones sound different from regular touch tones). Does anyone have any ideas what is causing any of the things I described above? I could understand this type of thing with the old mechanical switches, but since ESS has no moving parts, it seems to me it should either always work or not work at all. I don't understand how things can work just part of the time, or work different at different times. In real life: Frank Kienast Well: well!fgk@ucbvax.Berkeley.EDU CIS: 73327,3073 V-mail: 804-980-3733
cmr%carp.uucp@eecs.nwu.edu (Charles Rader) (06/27/89)
Some of the problems mentioned in Telecom 9.213 sound like an overloaded switch. When Michigan Bell first installed a new Northern Telecom switch in Southfield (major suburban office and commercial area) it was quite common for subscribers to wait several minutes before receiving a dialtone. Lack of response to call forwarding requests and inconsistencies in how many digits are dialed before a recording or reorder tone could also result from a switch that isn't responding fast enough. The variations in recordings could be caused by programming mistakes. Michigan Bell ultimately had to upgrade the switch as well as correct whatever programming bugs were present. It took many months before it stabilized.
john@decwrl.dec.com (John Higdon) (06/27/89)
In article <telecom-v09i0213m01@vector.dallas.tx.us>, well!fgk@lll-crg.llnl. gov (Frank G Kienast) writes: > I am curious as to how an exchange with ESS can have so many bugs and quirks > in it. I have noticed the following over the past couple years since we got > ESS. Each of the things described below has happened to me at least several > times. Gee, I thought at first you were describing my office! However, being an old hand at living with a clapped-out #1 ESS (NOT 1A!!!) maybe I can shed a little light. The switchmen there consider me their #1 PEST, but I feel entitled to bitch with my 10 lines! > Sometimes I pick up the phone and their is no dialtone. The touch tone pad > works, so there is a voltage, but no tone. If I hang up and retry several This can be one of two things. The most likely is that the office is simply overloaded. The 1E processor is much less adept at handling calls than the 1A, and since this is, after all, a common-control system, a lot of people going off-hook and dialing is going to make other people wait. Of a smaller possibility: the office has crashed! They're famous for that. > Sometimes in the middle of a call (both local and long distance), I get > mysteriously disconnected. All of a sudden there is a click and then > silence, but the voltage is still there. The party on the other end notes > the same thing. In all fairness, this could be occuring outside of the office in either the trunk circuits or your LD carrier. This happened to me the other night; I had no dial tone for fifteen minutes and when it came back all of my call forwarding on three lines had been cancelled. This was the processor crashing. A "reboot" clears tmp memory which is where call forwarding is stored. > Occasionally, the system won't let me program in call forwarding. I dial > 72# and get a fast busy (reorder). I dial 73# to make sure I'm not already This is defined as "temporary memory failure". Very common. Generally, it sounds like this: dial 72# [dial tone]. Dial number [two dial tone bursts then immediate busy signal (60 ipm)]. Dial 72# [dial tone]. Dial number [immediate reorder (120 ipm)]. I've heard your way, also. > Sometimes when you dial three digits that are not a local exchange, you get > a recording right away. Other times, you don't get the recording until you > dial seven digits. Also, no distinction is made between a number that is > invalid and one that is valid but requires a 1 first. In fact, the > recordings are used interchangably, with the "you must first dial a 1 when > calling this number" being used for a few weeks for both, then "your call > cannot being completed as dialed" being used for several weeks. This is just out and out sloppiness in your CO. Small telcos usually are plagued with this, but not BOCs. A "1" is not required in my exchange for long distance, so I've never heard that particular recording. > When you call an AT&T LD operator (by dialing 10288 0#, for example), you > hear four DTMF tones followed by a loud "clunk" followed by about ten more > DTMF tones (these tones sound different from regular touch tones). This is very surprising. It is identical to what you hear when dialing "0+" on a CONTAC equipped Xbar exchange. I can't believe that they glued CONTAC into an ESS office. Even a 1ESS can handle equal access in its generic programming. I'd love to hear that one for myself. > Does anyone have any ideas what is causing any of the things I described > above? I could understand this type of thing with the old mechanical > switches, but since ESS has no moving parts, it seems to me it should either > always work or not work at all. I don't understand how things can work just > part of the time, or work different at different times. Oh, but there are moving parts! The 1/1A ESS is simply an electonically controlled mechanical office. The actual switching is done with little bi-stable relays. Apply current in one direction and they close. Apply current in the other and they open. Why do you think it goes "kaplunk-klunk" when you get call-wasted? It disconnects you from your party, connects you momentarily to the call-wasting tone, then reconnects you to your party through a "conferencing" path. That's why the second beep (klunk) is less violent: you're already connected to the conferencing port. If you don't answer the second call and they hang up, it goes kaplunk. That's you and your party being moved back to the normal talk-path. -- John Higdon | P. O. Box 7648 | +1 408 723 1395 john@zygot.uucp | San Jose, CA 95150 | M o o !
jjjs@cbnewsc.att.com (james.j.sowa) (06/27/89)
In article <telecom-v09i0213m01@vector.dallas.tx.us> well!fgk@lll-crg.llnl.gov (Frank G Kienast) writes: >X-TELECOM-Digest: volume 9, issue 213, message 1 of 5 >Occasionally, the system won't let me program in call forwarding. I dial >72# and get a fast busy (reorder). I dial 73# to make sure I'm not already >forwarding, and get the reorder here too. I try hanging up, making other >calls first, etc. and forwarding still won't work. Then an hour or so later >it cures itself. Frank, I though the call fowarding activate and deactivate codes were *72.... and *73 respectively. Is this just a typo or are they really different on your local switch. --- ============================================================ Jim Sowa, att!cbnewsc!jjjs (312)979-4817 These views are not necessarily those of my employer
fgk@lll-crg.llnl.gov (Frank G Kienast) (07/02/89)
Thanks for all the responses to my questions about quirks in my ESS exchange. In addition to responses to this group, I have received nearly a dozen mail responses. The following is a summary of responses so far. Sometimes not receiving a dialtone is caused by the fact that the switch scans the subscriber loops. A bug in the software occasionally causes it to go off into oblivion until it times out and resets. (No explanation yet for the sequence of ascending tones that are sometimes heard just before receiving a dialtone.) Getting disconnected in the middle of a call (which I forgot to mention happens on local calls as well as long distance) could be caused by the failure of the (mechanical) relays which are driven by the ESS software (I had thought before that electronic switches were used). Inability on occasion to program in call forwarding is caused by a temporary lack of available memory, which is dynamically allocated when the request is made. Someone asked me if the 72# and 73# codes were a typo (they thought they were *72 and *73). I checked and 72# and 73# are indeed correct, but I tried *72 and *73 and these work also. Apparently they are interchangable. Sometimes receiving a recording after dialing three invalid digits, and on other occasions not receiving any response until seven digits are dialed may result from changes to the on-line database which are constantly being made due to trunk changes, the addition of new exchanges, etc. ESS screens each digit as it is dialed, so the capability of detecting an invalid sequence before seven digits are dialed is built in. The mixup between the "You must first dial a 1" and the "Your call cannot be completed as dialed" recordings is apparently a programming mistake. (The software comes with the switch, right? Wouldn't all switches of the same type and company have this problem then?) Getting different recordings after the first recording times out is also a programming problem (Hmm.. maybe they used GOTO's to branch to the routines that play the various recordings and forgot the jump out between them :-)) The tones heard when connecting to an AT&T operator are MF (Multi-Frequency) tones which are used by the network for routing calls. They are combinations of 700-900-1100-1300-1500-1700 Hz. Apparently intermediate routing is required because my area is served by a small telephone company that is not a former part of Bell. In response to questions I have received: I am in the 804-979 exchange. The local phone company is Centel. I believe they have a Northern Telecom switch. Finally, I'm wondering if someone could recommend a good book on ESS. I have only a very basic knowledge of telephone switching systems, but have a good understanding of programming (from high level languages down to assembly language) and electronics. I'm looking for such things as the history of ESS systems, problems that were encountered in their design and how they were overcome, information on the hardware and software typically used (type of processor and operating system, some sample flow charts or code, etc.), a listing of the various ESS versions (#1,1A, etc.) and their specifications, etc. Please let me know if you have any recommendations on what book(s) to read and where they can be obtained. In real life: Frank Kienast Well: well!fgk@ucbvax.Berkeley.EDU CIS: 73327,3073 V-mail: 804-980-3733
chuckh@uunet.uu.net (07/10/89)
I tried to mail this but it bounced, and since others may be interested, here it is. well!fgk@ucbvax.Berkely.EDU asked for books on ESS. These are some of the better articles I know of: Number 5ESS System is well written up in: AT&T Technical Journal Vol 64 Jul-Aug 1985, No 6, Part 2 about 250 pages all about the 5ESS system. Older AT&T Tech Journals (and Bell System Tech Journal, and Bell Labs Tech Journal, the name keeps changing :-) ) had similar material on the 1ESS, the 4ESS. I think the No 1 ESS was written up in BSTJ 43, No 5 (Sep 1964) I highly recommend the 1ESS issue. The processor used is very strange and entertaining! No 10A Remote Switching System (very interesting) BSTJ 61, No 4 (April 82) No 4 ESS BSTJ 56 NO 7 (Sep 1977) also (BSTJ 60, No 6 (Jul-Aug 1981) All of these should be available in larger city libraries and certainly any university library. They are in at least three different ones in Seattle, so you should be able to locate them. These will give you a good starting point, just follow the references. Chuck Huffington Apex Computer Company (206) 867-1900 uunet!apex!chuckh