[Mailman-Users] Mailman issue downloading full raw archive

[Mark Sapiro]

Jared Nyland wrote:
>
>Great Thanks for the help on this issue.  I have added that code to my
>private.py and now I am getting this error.
>
>
>Traceback:
>
>Traceback (most recent call last):
>  File "/usr/lib/mailman/scripts/driver", line 88, in run_main
>    sys.__stdout__.write(tempstdout.getvalue())
>MemoryError


You can try the following change to /usr/lib/mailman/scripts/driver.

locate the code

        try:
            try:
                sys.stderr = logger
                sys.stdout = tempstdout
                main()
                sys.__stdout__.write(tempstdout.getvalue())
            finally:
                sys.stderr = sys.__stderr__
                sys.stdout = sys.__stdout__

and change it to

        try:
            try:
                sys.stderr = logger
                sys.stdout = tempstdout
                main()
                tempstdout.seek(0)
                line = tempstdout.readline()
                while line:
                    sys.__stdout__.write(line)
                    line = tempstdout.readline()
            finally:
                sys.stderr = sys.__stderr__
                sys.stdout = sys.__stdout__

This and the prior change are really only a stopgap. I now see the
issue is twofold. The output of the CGI is written to an in-memory
string and then ultimately copied to the real stdout. The base code
actually copies things in a way that at times there are two full
copies of the file in memory at once. These changes do the copy a line
at a time so that there is only one file copy in memory plus a line
which should avoid the problem for now, but eventually, the file will
grow large enough so that even one copy won't fit in memory, but
before that, perhaps the browsers receiving this file will choke.

-- 
Mark Sapiro <msapiro at value.net>       The highway is for gamblers,
San Francisco Bay Area, California    better use your sense - B. Dylan