-
Notifications
You must be signed in to change notification settings - Fork 16
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Minimal Working Example Errors #13
Comments
Dear Michael,
I suggest you use our evaluation script (the *.py) to set up the Crane
instances on different machines and run them. My student Cheng Wang (cc'ed)
will help you set them up and answer your question.
…On Tue, Apr 24, 2018 at 12:11 PM, Michael Whittaker < ***@***.***> wrote:
Hello! I've written a very simple echo server
<https://github.com/mwhittaker/crane/blob/tweaks/apps/echo/server.py> in
Python and am trying to replicate the server using crane. I've read through
the instructions in the README which led me to read new-run.sh,
worker-run.py, worker-build.py, master.py, etc. I couldn't run those
scripts out of the box, so I tried to write the simplest script I could to
replicate the server without them.
I want to run with three servers (running on 127.0.0.1:7000-7002) and
three proxies (running on 127.0.0.1:9000-9002). For simplicity, I want to
run all three servers and all three proxies on the same machine with Parrot
but without LXC. I have the following two config files, which (I hope) set
this up properly:
- local.options
- nodes.local.cfg
I run this command three times to run the three proxies. This command
succeeds.
libevent_paxos/target/server.out -n $i -r -m s -c nodes.local.cfg -l ./proxy_logs
Then, I ran this to start the servers:
xtern/scripts/wrap-xtern.sh python server.py 700$i
Unfortunately, this does not work. Running this prints
terminate called after throwing an instance of 'boost::interprocess::interprocess_exception'
what(): No such file or directory
I read through the code and found that the first second line of code in
paxq_open_shared_mem inside of paxos-op-queue.cpp is throwing the
exception. It looks like there are some files that it expects to exist
inside of /dev/shm that don't exist. Do I have to make some files first
before running my server? I noticed the paxq_create_shared_mem function
above, but it is never called in the code. I added a call to it myself and
then the server runs! But, unfortunately when I try to connect a client to
a proxy, the messages are not being delivered.
Long story short, I want to replicate a simple server with crane as simply
as possible and am running into some troubles. Any help or guidance would
be greatly appreciated!
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#13>, or mute the thread
<https://github.com/notifications/unsubscribe-auth/AC8Q7OdwQkli6mrfdEOm7_l6y5tqXrEGks5trqXpgaJpZM4Tg-YP>
.
|
Hi @mwhittaker, the In a word, as mentioned by Dr.Heming, we recommend you to use our automatic scripts, otherwise there might be some weird problems that are difficult to debug. |
Ah, I see.
Okay, I'll give that a shot! Thanks for the quick help @hemingcui and @wangchenghku! |
I've set everything up to run on three EC2 servers with user name
where
which returns the following error message:
which is the same one I was experiencing before. @wangchenghku, how should I go about fixing this? You said earlier "paxq_create_shared_mem will only invoked by proxy when you have specified sched_with_dmt in the configuration file", but I do have |
Hello! I've written a very simple echo server in Python and am trying to replicate the server using crane. I've read through the instructions in the README which led me to read
new-run.sh
,worker-run.py
,worker-build.py
,master.py
, etc. I couldn't run those scripts out of the box, so I tried to write the simplest script I could to replicate the server without them.I want to run with three servers (running on
127.0.0.1:7000-7002
) and three proxies (running on127.0.0.1:9000-9002
). For simplicity, I want to run all three servers and all three proxies on the same machine with Parrot but without LXC. I have the following two config files, which (I hope) set this up properly:local.options
nodes.local.cfg
I run this command three times to run the three proxies. This command succeeds.
libevent_paxos/target/server.out -n $i -r -m s -c nodes.local.cfg -l ./proxy_logs
Then, I ran this to start the servers:
Unfortunately, this does not work. Running this prints
I read through the code and found that the first second line of code in
paxq_open_shared_mem
inside ofpaxos-op-queue.cpp
is throwing the exception. It looks like there are some files that it expects to exist inside of/dev/shm
that don't exist. Do I have to make some files first before running my server? I noticed thepaxq_create_shared_mem
function above, but it is never called in the code. I added a call to it myself and then the server runs! But, unfortunately when I try to connect a client to a proxy, the messages are not being delivered.Long story short, I want to replicate a simple server with crane as simply as possible and am running into some troubles. Any help or guidance would be greatly appreciated!
The text was updated successfully, but these errors were encountered: