# Attempt to reboot a node in debug state.
-import plc
+from monitor import const
+from monitor.database.info.model import *
+from monitor.wrapper import plc
api = plc.getAuthAPI()
import sys
import os
-import const
from getsshkeys import SSHKnownHosts
import subprocess
import time
-import database
-import moncommands
+from monitor.util import command as moncommands
from sets import Set
-import ssh.pxssh as pxssh
-import ssh.fdpexpect as fdpexpect
-import ssh.pexpect as pexpect
-from unified_model import *
-from emailTxt import mailtxt
+from pcucontrol.transports.ssh import pxssh as pxssh
+from pcucontrol.transports.ssh import fdpexpect as fdpexpect
+from pcucontrol.transports.ssh import pexpect as pexpect
+from monitor.model import *
+from monitor.wrapper.emailTxt import mailtxt
from nodeconfig import network_config_to_str
import traceback
-import monitorconfig
+from monitor import config
import signal
class Sopen(subprocess.Popen):
from Rpyc.Utils import *
fb = None
-def get_fbnode(node):
- global fb
- if fb is None:
- fb = database.dbLoad("findbad")
- fbnode = fb['nodes'][node]['values']
- return fbnode
-
class NodeConnection:
def __init__(self, connection, node, config):
self.node = node
args['port'] = self.port
args['user'] = 'root'
args['hostname'] = self.node
- args['monitordir'] = monitorconfig.MONITOR_SCRIPT_ROOT
+ args['monitordir'] = config.MONITOR_SCRIPT_ROOT
ssh_port = 22
if self.nosetup:
# NOTE: Nothing works if the bootcd is REALLY old.
# So, this is the first step.
- fbnode = get_fbnode(hostname)
+ fbnode = FindbadNodeRecord.get_latest_by(hostname=hostname).to_dict()
if fbnode['category'] == "OLDBOOTCD":
print "...NOTIFY OWNER TO UPDATE BOOTCD!!!"
args = {}
mailtxt.newbootcd_one[1] % args, True, db='bootcd_persistmessages')
loginbase = plc.siteId(hostname)
- m.send([const.PIEMAIL % loginbase, const.TECHEMAIL % loginbase])
+ emails = plc.getTechEmails(loginbase)
+ m.send(emails)
print "\tDisabling %s due to out-of-date BOOTCD" % hostname
api.UpdateNode(hostname, {'boot_state' : 'disable'})
mailtxt.baddisk[1] % args, True, db='hardware_persistmessages')
loginbase = plc.siteId(hostname)
- m.send([const.PIEMAIL % loginbase, const.TECHEMAIL % loginbase])
+ emails = plc.getTechEmails(loginbase)
+ m.send(emails)
conn.set_nodestate('disable')
return False
('nodehostname' , 'Configured node hostname does not resolve'),
('implementerror', 'Implementation Error'),
('readonlyfs' , '[Errno 30] Read-only file system'),
+ ('baddisk' , "IOError: [Errno 13] Permission denied: '/tmp/mnt/sysimg//vservers/\w+/etc/hosts'"),
('noinstall' , 'notinstalled'),
('bziperror' , 'bzip2: Data integrity error when decompressing.'),
('noblockdev' , "No block devices detected."),
('hardwarerequirefail' , 'Hardware requirements not met'),
('mkfsfail' , 'while running: Running mkfs.ext2 -q -m 0 -j /dev/planetlab/vservers failed'),
('nofilereference', "No such file or directory: '/tmp/mnt/sysimg//vservers/.vref/planetlab-f8-i386/etc/hosts'"),
+ ('kernelcopyfail', "cp: cannot stat `/tmp/mnt/sysimg/boot/kernel-boot': No such file or directory"),
('chrootfail' , 'Running chroot /tmp/mnt/sysimg'),
('modulefail' , 'Unable to get list of system modules'),
('writeerror' , 'write error: No space left on device'),
"bminit-cfg-auth-getplc-update-installinit-validate-rebuildinitrd-netcfg-update3-implementerror-nofilereference-update-debug-done",
"bminit-cfg-auth-getplc-update-hardware-installinit-installdisk-exception-mkfsfail-update-debug-done",
"bminit-cfg-auth-getplc-installinit-validate-rebuildinitrd-exception-chrootfail-update-debug-done",
+ "bminit-cfg-auth-getplc-update-installinit-validate-rebuildinitrd-netcfg-disk-update4-exception-chrootfail-update-debug-done",
+ "bminit-cfg-auth-getplc-update-hardware-installinit-installdisk-installbootfs-installcfg-installstop-update-installinit-validate-rebuildinitrd-netcfg-disk-update4-update3-update3-kernelcopyfail-exception-update-debug-done",
+ "bminit-cfg-auth-getplc-hardware-installinit-installdisk-installbootfs-installcfg-installstop-update-installinit-validate-rebuildinitrd-netcfg-disk-update4-update3-update3-kernelcopyfail-exception-update-debug-done",
"bminit-cfg-auth-getplc-installinit-validate-exception-noinstall-update-debug-done",
+ # actual solution appears to involve removing the bad files, and
+ # continually trying to boot the node.
+ "bminit-cfg-auth-getplc-update-installinit-validate-rebuildinitrd-netcfg-disk-update4-update3-update3-implementerror-update-debug-done",
]:
sequences.update({n : "restart_bootmanager_rins"})
"bminit-cfg-auth-implementerror-bootcheckfail-update-implementerror-bootupdatefail-done",
"bminit-cfg-auth-getplc-update-installinit-validate-rebuildinitrd-netcfg-update3-implementerror-nospace-update-debug-done",
"bminit-cfg-auth-getplc-hardware-installinit-installdisk-installbootfs-exception-downloadfail-update-debug-done",
+ "bminit-cfg-auth-getplc-update-installinit-validate-implementerror-update-debug-done",
]:
sequences.update({n: "restart_node_boot"})
m = PersistMessage(hostname, mailtxt.unknownsequence[0] % args,
mailtxt.unknownsequence[1] % args, False, db='unknown_persistmessages')
m.reset()
- m.send(['monitor-list@lists.planet-lab.org'])
+ m.send([config.cc_email])
conn.restart_bootmanager('boot')
m = PersistMessage(hostname, "Suspicous error from BootManager on %s" % args,
mailtxt.unknownsequence[1] % args, False, db='suspect_persistmessages')
m.reset()
- m.send(['monitor-list@lists.planet-lab.org'])
+ m.send([config.cc_email])
conn.restart_bootmanager('boot')
m = PersistMessage(hostname, mailtxt.plnode_cfg[0] % args, mailtxt.plnode_cfg[1] % args,
True, db='nodeid_persistmessages')
loginbase = plc.siteId(hostname)
- m.send([const.PIEMAIL % loginbase, const.TECHEMAIL % loginbase])
+ emails = plc.getTechEmails(loginbase)
+ m.send(emails)
conn.dump_plconf_file()
conn.set_nodestate('disable')
m = PersistMessage(hostname, mailtxt.plnode_network[0] % args, mailtxt.plnode_cfg[1] % args,
True, db='nodenet_persistmessages')
loginbase = plc.siteId(hostname)
- m.send([const.PIEMAIL % loginbase, const.TECHEMAIL % loginbase])
+ emails = plc.getTechEmails(loginbase)
+ m.send(emails)
conn.dump_plconf_file()
conn.set_nodestate('disable')
mailtxt.newalphacd_one[1] % args, True, db='bootcd_persistmessages')
loginbase = plc.siteId(hostname)
- m.send([const.PIEMAIL % loginbase, const.TECHEMAIL % loginbase])
+ emails = plc.getTechEmails(loginbase)
+ m.send(emails)
print "\tDisabling %s due to out-of-date BOOTCD" % hostname
conn.set_nodestate('disable')
mailtxt.baddisk[1] % args, True, db='hardware_persistmessages')
loginbase = plc.siteId(hostname)
- m.send([const.PIEMAIL % loginbase, const.TECHEMAIL % loginbase])
+ emails = plc.getTechEmails(loginbase)
+ m.send(emails)
conn.set_nodestate('disable')
elif sequences[s] == "update_hardware_email":
mailtxt.minimalhardware[1] % args, True, db='minhardware_persistmessages')
loginbase = plc.siteId(hostname)
- m.send([const.PIEMAIL % loginbase, const.TECHEMAIL % loginbase])
+ emails = plc.getTechEmails(loginbase)
+ m.send(emails)
conn.set_nodestate('disable')
elif sequences[s] == "bad_dns_email":
mailtxt.baddns[1] % args, True, db='baddns_persistmessages')
loginbase = plc.siteId(hostname)
- m.send([const.PIEMAIL % loginbase, const.TECHEMAIL % loginbase])
+ emails = plc.getTechEmails(loginbase)
+ m.send(emails)
conn.set_nodestate('disable')
if flag_set:
# MAIN -------------------------------------------------------------------
def main():
- import parser as parsermodule
+ from monitor import parser as parsermodule
parser = parsermodule.getParser()
parser.set_defaults(child=False, collect=False, nosetup=False, verbose=False,
git://git.onelab.eu / monitor.git / blobdiff