-
There appears to be a fundamental flaw in the design of the initialization
within libdmtcp.
The current design is this:
```
dmtcp_initialize() // Single point of entry for initialization
if…
-
So I have a process that writes to a file periodically and after checkpointing, the process continues running and continues to write to that file. Is there an option or way to have dmtcp kill the proc…
-
Currently, for a pre-existing socket, dmtcp treats it as a normal socket before checkpoint (i.e., dmtcp still performs leader election, draining, etc.), and as a non-shared socket on restart. This wor…
-
I get the following error on my system after I updated to libc 2.25 recently:
```
[41000] restore_libc.c:511 in TLSInfo_VerifyPidTid; REASON= ERROR: getpid(19559), tls pid(19553), and tls tid(1955…
-
This description of the race condition can be improved, and an automated gdbinit file can be created to reproduce this bug. But for now, I don't have the time. So, here is an informal description of…
-
This plugin can be used by a user to specify what ckpt/restart behavior he/she wants for different open files. To this end I propose the following changes:
- The plugin should read and act according t…
-
Right now the DMTCP code base doesn't have a uniform coding style and it is often counter productive. Let's settle down on one of the existing opensource coding standards and enforce it in the code b…
-
DMTCP promotes pipes to socketpairs. However, this may break some of the semantics that applications may depend upon. We need to introduce a new "PIPE" type to the IPC plugin.
The issue was originall…
-
When using dmtcp with python, I am finding that a particular scipy module import causes the launch to hang.
Here's the smallest example that illustrates it:
```
from __future__ import print_functio…
ghost updated
5 years ago
-
I have got the latest version of DMTCP installed and tested with a simple python script to make sure it was checkpointing and restoring fine. I use SGE for job queuing/scheduling. the issue i am runni…