Error Messages

Q: I am getting "inconsistent license key" error messages when starting up lmgrd. Why? What should I do?

A: This happens when the encryption code that avantd calculates is different from the encryption key used to generate the license key. The following pieces of information from the license key are used: the host id of the server, the software expiration date, the increment name, and the software version.

If the user changes any of these fields, you will get an inconsistent license key error message. This can also happen when you are using different versions of avantd and lmgrd; however, it is very unlikely that this will cause a problem.
Avant! does not recommend using lmreread to start a new license. Shut the license server down using:

lmdown -c /path/to/license/file

and then start the new license server in the normal manner.

Solution: 134

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Q: I am getting a TCL error: "Cannot find config.status" when running the $installdir/bin/config script to configure Star-Hspice/AvanWaves. Why?

A: Your installation is probably fine. No matter what products you choose to install, the configuration script tries to display the config.status file. Unfortunately, not all installation options generate a "config.status" file.

You can verify that the configuration was successful by reading the automatically generated logfile.This is placed in /tmp/avanti_####/config.log, where #### are numbers specific to the date and time of the installation.

Solution: 137

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Q: I am getting the following error message:

** error**: iob_loads2:9612: Number of input voltage sources in IOB = 0

What does this mean?

A: Certain connections in IBIS elements must be connected to ideal sources (e.g., output buffers must have their input node connected to an ideal source). For more information on which nodes must be connected, please see Using IBIS Models, in the Star-Hspice Manual .

Solution: 145

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Q: Why is the simulate button grayed out in HSPUI, even if a simulation is not being performed?

A: This is usually caused by a computer crash during a simulation. To ungray the simulate button, type in a command prompt:

"del %tmp%\job.run"

Alternatively, you search for the file job.run , and delete it.

Solution: 293

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Q: Why do I get the error " inductor/voltage source loop "?

A: This is normally caused by:

1. A voltage source with the inductor connected to the same point.

2. A voltage source with an inductor connected directly across its nodes.

3. Two or more inductors connected in a loop with nothing to limit the current.

The best way to solve this type of error is to connect a small series resistance (1 nano ohm or smaller usually works).

Solution: 464

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Q: I get the error message: " Time: command terminated abnormally ." Why?

A: This means Star-Hspice stops unexpectedly, which can be due to a number of problems (such as convergence problems or a numerical overflow). Looking at the *.lis and *.st0 files can usually give clues to what happened. Also, try to find out if the simulation concluded using a different version of Star-Hspice or on other platforms. If it did, it could be a bug. Otherwise, it is most likely a problem with the designs or model cards.

The message, " Time: command terminated abnormally " is output by the UNIX utility TIME, not Star-Hspice itself. The TIME utility reports various system use information on the processes it monitors. When a process being monitored stops unexpectedly, TIME issues this warning message.

Solution: 520

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Q: Why am I getting the following error in my license.log file when I try to start the avantd daemon:

ATTENTION ... "avantd" vendor daemon CAN NOT co-exist with monarc lockfile * error (-1) with "monarc", program aborted... Please correct problem and restart daemons

A: This is caused by an earlier Avanti vendor daemon. Look in the /usr/tmp directory for a file named either lockmonarcd or lock.monarcd. Delete the lockfile and restart lmgrd. The avantd daemon will now be able to start.

Solution: 528

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Q: Why do I get a Star-Hspice and AvanWaves " configure unsuccessful " error?

A: If $installdir is previously set, when you install a new version of Star-Hspice the path for the configuration will not be correct (it will try to install over your previous installation). This causes a " configure unsuccessful " error. To fix this, please set the correct path to your install directory in the configure utility.

Solution: 561

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Q: Why do I get the error message: " **error**: system file descriptors may be used up "?

A: This is an error reported by your operating system. It happens when Star-Hspice tries to open more files than is permitted. If you have a large number of .ALTER statements or if you read in lots of files (.LIB or .INCLUDE), you will have to raise your system's limit.

With sh/ksh/bash, you can use the "ulimit" built-in to display or change the limits for the current shell and all the processes you launch from this shell. Use:

$ ulimit -n

to display the current limit. For example, on my system, I get 64; this means that any process I launch from this shell cannot open more than 64 files at once. To raise this limit to 256, for instance, type:

$ ulimit -n 256

There is a "hard" limit that you cannot exceed (although the root can change the hard limit). To see what the hard limit is on your system, type:

$ ulimit -Hn

On csh/tcsh, use the "limit" built-in do this (type limit to see a list of limits). For example, on my system:

cputime unlimited

filesize unlimited

datasize 2097148 kbytes

stacksize 8192 kbytes

coredumpsize unlimited

descriptors 64

memorysize unlimited

To increase the number of open files (file descriptors) to 256, type:

$ limit descriptors 256

To view the hard limits, type:

$ limit -h

Solution: 613

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Q: Why do I get this message: "**error ** reference 0:nch not found " even when NCH appears in my library?

A: Star-Hspice has a feature called "automatic model selection," which is probably used in your library. You can identify models that use this feature if the model name is followed by a period and a suffix, as in:

.MODEL nch.1 nmos ...

It is likely that you have multiple .model nch.* statements that define the model's parameters for different sizes of the transistor. These models use LMIN, LMAX, WMIN and WMAX to determine which model applies to a given transistor dimensions.
For instance, if model nch.1 includes LMIN=0.15u LMAX=0.25u WMIN=0.15u WMAX=1.0u, then the model is only applied to transistors with lengths between 0.15um and 0.25um and widths between 0.15um and 1.0um.
If you are getting a " reference not found " error, then your transistor dimensions (L and W) do not fall in the ranges specified by any of the models in your library.

Solution: 642

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Star-Hspice Manual - Release 2001.2 - June 2001