Can't run rake maglev:restart or maglev:start

[zenspider]

This seems to be true for a fresh checkout. My only way of restarting a stone right now is to run update.sh, which is hella time consuming considering I only want to reload my parser lib into the stone.

 844 % rake maglev:restart
=== Starting with netldiname gs64ldi
startstone[Info]: GemStone version '3.1.0.2.1'
startstone[Info]: Starting Stone repository monitor 'maglev'.
startstone[Info]: GEMSTONE is: '/home/ryan/maglev/gemstone'.
startstone[Info]:
    GEMSTONE_SYS_CONF=/home/ryan/maglev/etc/conf.d/maglev.conf
    GEMSTONE_EXE_CONF=/home/ryan/maglev/maglev.conf
startstone[Info]: Log file is '/home/ryan/maglev/log/maglev/maglev.log'.

startstone[Error]: Stone process (id=21753) has died.
startstone[Error]: Examine '/home/ryan/maglev/log/maglev/maglev.log' for more information.  Excerpt follows:
STN_TRAN_LOG_DEBUG_LEVEL = 0;
STN_TRAN_LOG_DIRECTORIES = "/home/ryan/maglev/data/maglev/tranlog",
"/home/ryan/maglev/data/maglev/tranlog";
STN_TRAN_LOG_LIMIT = 1000;
STN_TRAN_LOG_PREFIX = "tranlog";
STN_TRAN_LOG_SIZES = 1000,
1000;
STN_TRAN_Q_TO_RUN_Q_THRESHOLD = 6;
STN_WELL_KNOWN_PORT_NUMBER = 0;

--- 04/08/15 17:19:06 PDT ---
 _____________________________________________________________________________
|     ERROR: Unable to stat() key file /home/ryan/maglev/versions/GemStone-34619.Linux-x86_64/sys/gemstone.key:
| errno=2,ENOENT, The file or directory specified cannot be found             |
|_____________________________________________________________________________|
rake aborted!
Command failed with status (3): [env GEMSTONE_LOGDIR=/home/ryan/maglev/log/...]
/home/ryan/maglev/rakelib/contrib/ottobehrens/gemstone_installation.rb:45:in `gs_sh'
/home/ryan/maglev/rakelib/contrib/ottobehrens/stone.rb:246:in `gs_sh'
/home/ryan/maglev/rakelib/contrib/ottobehrens/stone.rb:109:in `start'
/home/ryan/maglev/rakelib/maglev_stone.rb:59:in `start'
/home/ryan/maglev/rakelib/contrib/ottobehrens/stone.rb:125:in `restart'
/home/ryan/maglev/Rakefile:185:in `block in task_gemstone'
Tasks: TOP => maglev:restart
(See full trace by running task with --trace)

[AllenOtis]

I know nothing about rake and am not qualified to fix this.

If the underlying startstone command is failing when executed from
the shell as $GEMSTONE/bin/startstone
I will help fix those failures.

My advice for parser work is to start stone once , and load the ruby
bootstrap code.
Then you want to to directly execute topaz to run ruby to test
your parser changes. There is no need to stop/restart the stone
after installing a new libmagparse library . You may want to customize
some scripts. You want to be able to compile and link the libmagparse
and run individual ruby programs while leaving the stone running.
I don't use rake when I am doing ruby VM debugging ...

Allen

On Wed, Apr 8, 2015 at 5:21 PM, Ryan Davis [email protected] wrote:

This seems to be true for a fresh checkout. My only way of restarting a
stone right now is to run update.sh, which is hella time consuming
considering I only want to reload my parser lib into the stone.

844 % rake maglev:restart
=== Starting with netldiname gs64ldi
startstone[Info]: GemStone version '3.1.0.2.1'
startstone[Info]: Starting Stone repository monitor 'maglev'.
startstone[Info]: GEMSTONE is: '/home/ryan/maglev/gemstone'.
startstone[Info]:
GEMSTONE_SYS_CONF=/home/ryan/maglev/etc/conf.d/maglev.conf
GEMSTONE_EXE_CONF=/home/ryan/maglev/maglev.conf
startstone[Info]: Log file is '/home/ryan/maglev/log/maglev/maglev.log'.

startstone[Error]: Stone process (id=21753) has died.
startstone[Error]: Examine '/home/ryan/maglev/log/maglev/maglev.log' for more information. Excerpt follows:
STN_TRAN_LOG_DEBUG_LEVEL = 0;
STN_TRAN_LOG_DIRECTORIES = "/home/ryan/maglev/data/maglev/tranlog",
"/home/ryan/maglev/data/maglev/tranlog";
STN_TRAN_LOG_LIMIT = 1000;
STN_TRAN_LOG_PREFIX = "tranlog";
STN_TRAN_LOG_SIZES = 1000,
1000;
STN_TRAN_Q_TO_RUN_Q_THRESHOLD = 6;
STN_WELL_KNOWN_PORT_NUMBER = 0;

--- 04/08/15 17:19:06 PDT ---

---

| ERROR: Unable to stat() key file /home/ryan/maglev/versions/GemStone-34619.Linux-x86_64/sys/gemstone.key:
| errno=2,ENOENT, The file or directory specified cannot be found |
|_____________________________________________________________________________|
rake aborted!
Command failed with status (3): [env GEMSTONE_LOGDIR=/home/ryan/maglev/log/...]
/home/ryan/maglev/rakelib/contrib/ottobehrens/gemstone_installation.rb:45:in gs_sh' /home/ryan/maglev/rakelib/contrib/ottobehrens/stone.rb:246:ings_sh'
/home/ryan/maglev/rakelib/contrib/ottobehrens/stone.rb:109:in start' /home/ryan/maglev/rakelib/maglev_stone.rb:59:instart'
/home/ryan/maglev/rakelib/contrib/ottobehrens/stone.rb:125:in restart' /home/ryan/maglev/Rakefile:185:inblock in task_gemstone'
Tasks: TOP => maglev:restart
(See full trace by running task with --trace)

—
Reply to this email directly or view it on GitHub
#387.

[zenspider]

It's not the keyfile it is mentioning? I found one in a MUCH different location, but I wasn't sure what was up with that.

So, the libmagparse library isn't loaded by the server, only the client?

How do I directly execute topaz to run ruby? How is that different from bin/maglev-ruby?

[AllenOtis]

libmagparse is only loaded by the VM, i.e.
by the process started as
topaz -l
You are correct . bin/maglev-ruby is the way to run the
topaz process . But there should be no need to stop/restart
the stone each time you want to run ruby.
Your startstone did complain about lack of a key file.
I suspect the args to startstone from the maglev restart
script must be different than from the first install or startstone.
Or else the rake scripting is somehow deleting the keyfile
from it's initial install location.

On Wed, Apr 8, 2015 at 6:49 PM, Ryan Davis [email protected] wrote:

It's not the keyfile it is mentioning? I found one in a MUCH different
location, but I wasn't sure what was up with that.

So, the libmagparse library isn't loaded by the server, only the client?

How do I directly execute topaz to run ruby? How is that different from
bin/maglev-ruby?

—
Reply to this email directly or view it on GitHub
#387 (comment).

[johnnyt]

@zenspider - you should be able to use maglev start and maglev stop in place of the rake tasks