From 5c41a475f7d198274e46f2c5e6f033f06794288b Mon Sep 17 00:00:00 2001 From: Stephen Soltesz Date: Fri, 19 Jun 2009 01:44:06 +0000 Subject: [PATCH] added bootmanager log for monitor-collected log files to be referenced by web added keyboard interrupt exception to nodequery addressed root cause of IntegrityErrors from lousy code in controllers.py no elements of the Objects are assigned. this patch addresses functionality, need to clean up for clarity. adjusted templates to use new aggregate object of nodes, pcus, and 'pre-processed' values updated emailTxt to use just 'pattern' in www node urls added session.flush() to bootman.py to write out ActionsRecords --- Monitor.spec | 12 +- automate-default.sh | 1 + monitor-server.cron | 2 +- monitor-server.init | 1 + monitor/bootman.py | 44 +++-- monitor/wrapper/emailTxt.py | 10 +- monitor/wrapper/plccache.py | 14 +- nodequery.py | 7 +- web/MonitorWeb/monitorweb/controllers.py | 165 +++++++++--------- .../monitorweb/templates/node_template.kid | 22 +-- .../monitorweb/templates/nodelist.kid | 4 +- .../monitorweb/templates/pculist.kid | 16 +- .../monitorweb/templates/pcuview.kid | 52 +++--- 13 files changed, 190 insertions(+), 160 deletions(-) diff --git a/Monitor.spec b/Monitor.spec index c28b7de..c5578a6 100644 --- a/Monitor.spec +++ b/Monitor.spec @@ -6,7 +6,7 @@ %define name monitor %define version 3.0 -%define taglevel 16 +%define taglevel 15 %define release %{taglevel}%{?pldistro:.%{pldistro}}%{?date:.%{date}} %global python_sitearch %( python -c "from distutils.sysconfig import get_python_lib; print get_python_lib(1)" ) @@ -137,6 +137,7 @@ install -d $RPM_BUILD_ROOT/data/var/lib/%{name}/archive-pdb install -d $RPM_BUILD_ROOT/var/lib/%{name} install -d $RPM_BUILD_ROOT/var/lib/%{name}/archive-pdb install -d $RPM_BUILD_ROOT/var/www/cgi-bin/monitor/ +install -d $RPM_BUILD_ROOT/var/www/html/monitorlog/ install -D -m 755 monitor-server.init $RPM_BUILD_ROOT/%{_sysconfdir}/plc.d/monitor @@ -146,6 +147,7 @@ rsync -a --exclude www --exclude archive-pdb --exclude .svn --exclude CVS \ echo " * Installing web pages" rsync -a www/ $RPM_BUILD_ROOT/var/www/cgi-bin/monitor/ +rsync -a log/ $RPM_BUILD_ROOT/var/www/html/monitorlog/ echo " * Installing cron job for automated polling" install -D -m 644 monitor-server.cron $RPM_BUILD_ROOT/%{_sysconfdir}/cron.d/monitor-server.cron @@ -225,6 +227,8 @@ easy_install --build-directory /var/tmp -UZ Elixir # crazy openssl libs for racadm binary ln -s /lib/libssl.so.0.9.8b /usr/lib/libssl.so.2 +mkdir /usr/share/monitor/.ssh +chmod 700 /usr/share/monitor/.ssh if grep 'pam_loginuid.so' /etc/pam.d/crond ; then sed -i -e 's/^session required pam_loginuid.so/#session required pam_loginuid.so/g' /etc/pam.d/crond @@ -262,12 +266,6 @@ chkconfig --add monitor-runlevelagent chkconfig monitor-runlevelagent on %changelog -* Wed Jun 17 2009 Stephen Soltesz - Monitor-3.0-16 -- Added Rpyc from 1.0 branch. -- add pcuhistory -- add setup-agent for password protected keys. -- other minor improvements. - * Wed Jun 17 2009 Stephen Soltesz - Monitor-3.0-15 - automate install - auto-close tickets diff --git a/automate-default.sh b/automate-default.sh index 958d578..66a42d9 100755 --- a/automate-default.sh +++ b/automate-default.sh @@ -75,6 +75,7 @@ ps ax | grep locfg | grep -v grep | awk '{print $1}' | xargs -r kill || : ${MONITOR_SCRIPT_ROOT}/policy.py $DATE ${MONITOR_SCRIPT_ROOT}/checksync.py $DATE || : service plc restart monitor +curl -s 'http://summer.cs.princeton.edu/status/tabulator.cgi?table=table_nodeview&formatcsv' > /var/lib/monitor/comon/$DATE.comon.csv cp ${MONITOR_SCRIPT_ROOT}/monitor.log ${MONITOR_ARCHIVE_ROOT}/`date +%F-%H:%M`.monitor.log rm -f $MONITOR_PID diff --git a/monitor-server.cron b/monitor-server.cron index 4d02a82..73f0fdf 100644 --- a/monitor-server.cron +++ b/monitor-server.cron @@ -2,5 +2,5 @@ # automated actions for debug nodes. 01 * * * * root /usr/share/monitor/automate-default.sh 2>&1 > /usr/share/monitor/monitor.log -30 * * * * root /etc/plc.d/monitor sync 2>&1 >> /var/log/monitorsync.log +#30 * * * * root /etc/plc.d/monitor sync 2>&1 >> /var/log/monitorsync.log diff --git a/monitor-server.init b/monitor-server.init index 5ffe974..4b5bbb0 100644 --- a/monitor-server.init +++ b/monitor-server.init @@ -192,6 +192,7 @@ API_AUTH_PASSWORD=${PLC_MONITOR_DBPASSWORD} MONITOR_SCRIPT_ROOT=${MONITORPATH} MONITOR_DATA_ROOT=/var/lib/monitor MONITOR_ARCHIVE_ROOT=/var/lib/monitor/archive-pdb +MONITOR_BOOTMANAGER_LOG=/var/www/html/monitorlog MONITOR_HOSTNAME=${PLC_MONITOR_HOST} MONITOR_IP=${PLC_MONITOR_IP} diff --git a/monitor/bootman.py b/monitor/bootman.py index 2afbbf6..9fe28af 100755 --- a/monitor/bootman.py +++ b/monitor/bootman.py @@ -68,18 +68,34 @@ class NodeConnection: return "unknown" def get_dmesg(self): + t_stamp = time.strftime("%Y-%m-%d-%H:%M") self.c.modules.os.system("dmesg > /var/log/dmesg.bm.log") - download(self.c, "/var/log/dmesg.bm.log", "log/dmesg.%s.log" % self.node) - log = open("log/dmesg.%s.log" % self.node, 'r') + download(self.c, "/var/log/dmesg.bm.log", "%s/history/%s-dmesg.%s.log" % (config.MONITOR_BOOTMANAGER_LOG, t_stamp, self.node)) + os.system("cp %s/history/%s-dmesg.%s.log %s/dmesg.%s.log" % (config.MONITOR_BOOTMANAGER_LOG, t_stamp, self.node, config.MONITOR_BOOTMANAGER_LOG, self.node)) + log = open("%s/dmesg.%s.log" % (config.MONITOR_BOOTMANAGER_LOG, self.node), 'r') return log def get_bootmanager_log(self): - download(self.c, "/tmp/bm.log", "log/bm.%s.log.gz" % self.node) - #os.system("zcat log/bm.%s.log.gz > log/bm.%s.log" % (self.node, self.node)) - os.system("cp log/bm.%s.log.gz log/bm.%s.log" % (self.node, self.node)) - log = open("log/bm.%s.log" % self.node, 'r') + t_stamp = time.strftime("%Y-%m-%d-%H:%M") + download(self.c, "/tmp/bm.log", "%s/history/%s-bm.%s.log" % (config.MONITOR_BOOTMANAGER_LOG, t_stamp, self.node)) + os.system("cp %s/history/%s-bm.%s.log %s/bm.%s.log" % (config.MONITOR_BOOTMANAGER_LOG, t_stamp, self.node, config.MONITOR_BOOTMANAGER_LOG, self.node)) + log = open("%s/bm.%s.log" % (config.MONITOR_BOOTMANAGER_LOG, self.node), 'r') return log + +# def get_dmesg(self): +# self.c.modules.os.system("dmesg > /var/log/dmesg.bm.log") +# download(self.c, "/var/log/dmesg.bm.log", "log/dmesg.%s.log" % self.node) +# log = open("log/dmesg.%s.log" % self.node, 'r') +# return log +# +# def get_bootmanager_log(self): +# download(self.c, "/tmp/bm.log", "log/bm.%s.log.gz" % self.node) +# #os.system("zcat log/bm.%s.log.gz > log/bm.%s.log" % (self.node, self.node)) +# os.system("cp log/bm.%s.log.gz log/bm.%s.log" % (self.node, self.node)) +# log = open("log/bm.%s.log" % self.node, 'r') +# return log + def dump_plconf_file(self): c = self.c self.c.modules.sys.path.append("/tmp/source/") @@ -627,8 +643,12 @@ class DebugInterface: return sequence - def restore(sitehist, hostname, config=None, forced_action=None): + ret = restore_basic(sitehist, hostname, config, forced_action) + session.flush() + return ret + +def restore_basic(sitehist, hostname, config=None, forced_action=None): # NOTE: Nothing works if the bootcd is REALLY old. # So, this is the first step. @@ -650,16 +670,8 @@ def restore(sitehist, hostname, config=None, forced_action=None): debugnode = DebugInterface(hostname) conn = debugnode.getConnection() - #print "conn: %s" % conn - #print "trying to use conn after returning it." - #print conn.c.modules.sys.path - #print conn.c.modules.os.path.exists('/tmp/source') if type(conn) == type(False): return False - #if forced_action == "reboot": - # conn.restart_node('reinstall') - # return True - boot_state = conn.get_boot_state() if boot_state != "debug": print "... %s in %s state: skipping..." % (hostname , boot_state) @@ -689,7 +701,7 @@ def restore(sitehist, hostname, config=None, forced_action=None): print "...Should investigate. Skipping node." # TODO: send message related to these errors. - if not found_within(recent_actions, 'newbootcd_notice', 3): + if not found_within(recent_actions, 'baddisk_notice', 3): log=conn.get_dmesg().read() sitehist.sendMessage('baddisk_notice', hostname=hostname, log=log) diff --git a/monitor/wrapper/emailTxt.py b/monitor/wrapper/emailTxt.py index 77e8576..9b7b456 100644 --- a/monitor/wrapper/emailTxt.py +++ b/monitor/wrapper/emailTxt.py @@ -487,7 +487,7 @@ Thank you for your help, """ While trying to automatically recover this machine: - http://www.planet-lab.org/db/nodes/index.php?nodepattern=%(hostname)s + http://www.planet-lab.org/db/nodes/index.php?pattern=%(hostname)s We encountered an unknown situation. Please re-code to handle, or manually intervene to repair this host. @@ -517,7 +517,7 @@ It is essential that the AMT feature be configured to enable PlanetLab staff to Configure the DC7800 AMT feature : https://www.planet-lab.org/AMT Add a PCU to your site : https://www.planet-lab.org/db/sites/pcu.php Associate your node with the PCU : Follow the 'My Site' link - Finally, download the Boot Image : https://www.planet-lab.org/db/nodes/index.php?nodepattern=%(hostname)s + Finally, download the Boot Image : https://www.planet-lab.org/db/nodes/index.php?pattern=%(hostname)s Burn Boot Image to media & Reboot your node You can confirm that your machine's PCU is correctly configured by visiting the AMT @@ -573,7 +573,7 @@ Thank you for your help, """ While trying to automatically recover this machine: - http://www.planet-lab.org/db/nodes/index.php?nodepattern=%(hostname)s + http://www.planet-lab.org/db/nodes/index.php?pattern=%(hostname)s We encountered an failed hardware requirement. Please look at the log below to determine the exact nature of the failure, either Disk, CPU, Network, or Mimial RAM was not satisfied. @@ -637,7 +637,7 @@ Thank you for your help, As part of PlanetLab node monitoring, we noticed that %(hostname)s has a network configuration error related to DNS or hostname lookups. Often this can happen either due local configuraiton changes, or a misconfiguration of the node's DNS servers. To resolve the issue we require your assistance. All that is needed is to visit: - https://www.planet-lab.org/db/nodes/index.php?nodepattern=%(hostname)s + https://www.planet-lab.org/db/nodes/index.php?pattern=%(hostname)s Find the primary node network entry and confirm that the settings are correct. @@ -658,7 +658,7 @@ BootManager.log output follows: nodeconfig_notice=(""" Please Update Configuration file for PlanetLab node %(hostname)s""", """As part of PlanetLab node monitoring, we noticed %(hostname)s has an out-dated plnode.txt file with no NODE_ID or a mis-matched HOSTNAME. This can happen either due to an initial configuration failure at your site, with information entered into our database, or after a software upgrade. To resolve the issue we require your assistance. All that is needed is to visit: - https://www.planet-lab.org/db/nodes/index.php?nodepattern=%(hostname)s + https://www.planet-lab.org/db/nodes/index.php?pattern=%(hostname)s Then, select, "Download -> Download plnode.txt file for %(hostname)s" menu. This will generate a new configuration file for your node. Copy this file to the appropriate read-only media, either floppy or USB stick, and reboot the machine. diff --git a/monitor/wrapper/plccache.py b/monitor/wrapper/plccache.py index fea4c72..ac23f1b 100755 --- a/monitor/wrapper/plccache.py +++ b/monitor/wrapper/plccache.py @@ -63,8 +63,8 @@ plcdb_id2lb = None def init(): import traceback - print "IMPORTING PLCCACHE: ", - traceback.print_stack() + #print "IMPORTING PLCCACHE: ", + #traceback.print_stack() global l_sites global l_nodes global l_pcus @@ -73,14 +73,15 @@ def init(): global plcdb_id2lb print "initing plccache" + print "collecting plcsites" dbsites = PlcSite.query.all() l_sites = [ s.plc_site_stats for s in dbsites ] - print "plcnode" + print "collecting plcnodes" dbnodes = PlcNode.query.all() l_nodes = [ s.plc_node_stats for s in dbnodes ] - print "plcpcu" + print "collecting plcpcus" dbpcus = PlcPCU2.query.all() l_pcus = [] for s in dbpcus: @@ -90,11 +91,10 @@ def init(): 'model', 'password', 'ports']: pcu[k] = getattr(s, k) l_pcus.append(pcu) - #l_pcus = [ s.plc_pcu_stats for s in dbpcus ] - print "dsites_from_lsites" + print "building id2lb" (d_sites,id2lb) = dsites_from_lsites(l_sites) - print "dsn_from_dsln" + print "building lb2hn" (plcdb, hn2lb, lb2hn) = dsn_from_dsln(d_sites, id2lb, l_nodes) plcdb_hn2lb = hn2lb diff --git a/nodequery.py b/nodequery.py index 738e58d..2fe0e4a 100755 --- a/nodequery.py +++ b/nodequery.py @@ -317,6 +317,9 @@ def node_select(str_query, nodelist=None, fb=None): fb_noderec = None #fb_noderec = FindbadNodeRecord.query.filter(FindbadNodeRecord.hostname==node).order_by(FindbadNodeRecord.date_checked.desc()).first() fb_noderec = FindbadNodeRecord.get_latest_by(hostname=node) + except KeyboardInterrupt: + print "Exiting at user request: Ctrl-C" + sys.exit(1) except: print traceback.print_exc() continue @@ -328,7 +331,6 @@ def node_select(str_query, nodelist=None, fb=None): #if 'plcnode' in fb_nodeinfo: # fb_nodeinfo.update(fb_nodeinfo['plcnode']) - #if verifyDBrecord(dict_query, fb_nodeinfo): if verify(dict_query, fb_nodeinfo): #print fb_nodeinfo.keys() #print node #fb_nodeinfo @@ -414,6 +416,9 @@ def main(): fb_noderec = FindbadNodeRecord.get_latest_by(hostname=node) if not fb_noderec: continue fb_nodeinfo = fb_noderec.to_dict() + except KeyboardInterrupt: + print "Exiting at user request: Ctrl-C" + sys.exit(1) except: print traceback.print_exc() continue diff --git a/web/MonitorWeb/monitorweb/controllers.py b/web/MonitorWeb/monitorweb/controllers.py index aad1f21..dba81e9 100644 --- a/web/MonitorWeb/monitorweb/controllers.py +++ b/web/MonitorWeb/monitorweb/controllers.py @@ -22,6 +22,9 @@ from monitor.wrapper.plccache import plcdb_hn2lb as site_hn2lb from monitorweb.templates.links import * +# make it easier group objects without invoking the elixir auto-write feature. +class aggregate: pass + def query_to_dict(query): """ take a url query string and chop it up """ @@ -76,79 +79,85 @@ def format_pcu_shortstatus(pcu): return status def prep_pcu_for_display(pcu): + agg = aggregate() + agg.pcu = pcu try: - pcu.loginbase = PlcSite.query.get(pcu.plc_pcu_stats['site_id']).plc_site_stats['login_base'] + agg.loginbase = PlcSite.query.get(pcu.plc_pcu_stats['site_id']).plc_site_stats['login_base'] except: - pcu.loginbase = "unknown" + agg.loginbase = "unknown" - pcu.ports = format_ports(pcu.port_status, pcu.plc_pcu_stats['model']) - pcu.status = format_pcu_shortstatus(pcu) + agg.ports = format_ports(pcu.port_status, pcu.plc_pcu_stats['model']) + agg.status = format_pcu_shortstatus(pcu) #print pcu.entry_complete - pcu.entry_complete_str = pcu.entry_complete + agg.entry_complete_str = pcu.entry_complete #pcu.entry_complete_str += "".join([ f[0] for f in pcu.entry_complete.split() ]) if pcu.dns_status == "NOHOSTNAME": - pcu.dns_short_status = 'NoHost' + agg.dns_short_status = 'NoHost' elif pcu.dns_status == "DNS-OK": - pcu.dns_short_status = 'Ok' + agg.dns_short_status = 'Ok' elif pcu.dns_status == "DNS-NOENTRY": - pcu.dns_short_status = 'NoEntry' + agg.dns_short_status = 'NoEntry' elif pcu.dns_status == "NO-DNS-OR-IP": - pcu.dns_short_status = 'NoHostOrIP' + agg.dns_short_status = 'NoHostOrIP' elif pcu.dns_status == "DNS-MISMATCH": - pcu.dns_short_status = 'Mismatch' + agg.dns_short_status = 'Mismatch' + return agg class NodeWidget(widgets.Widget): pass def prep_node_for_display(node): + agg = aggregate() + agg.node = node + if node.plc_pcuid: pcu = FindbadPCURecord.get_latest_by(plc_pcuid=node.plc_pcuid) if pcu: - node.pcu_status = pcu.reboot_trial_status - node.pcu_short_status = format_pcu_shortstatus(pcu) - node.pcu = pcu - prep_pcu_for_display(node.pcu) + agg.pcu_status = pcu.reboot_trial_status + agg.pcu_short_status = format_pcu_shortstatus(pcu) + agg.pcu = prep_pcu_for_display(pcu) else: - node.pcu_short_status = "none" - node.pcu_status = "nodata" - node.pcu = None + agg.pcu_short_status = "none" + agg.pcu_status = "nodata" + agg.pcu = None else: - node.pcu_status = "nopcu" - node.pcu_short_status = "none" - node.pcu = None + agg.pcu_status = "nopcu" + agg.pcu_short_status = "none" + agg.pcu = None if node.kernel_version: - node.kernel = node.kernel_version.split()[2] + agg.kernel = node.kernel_version.split()[2] else: - node.kernel = "" + agg.kernel = "" try: - node.loginbase = PlcSite.query.get(node.plc_node_stats['site_id']).plc_site_stats['login_base'] + agg.loginbase = PlcSite.query.get(node.plc_node_stats['site_id']).plc_site_stats['login_base'] except: - node.loginbase = "unknown" + agg.loginbase = "unknown" - if node.loginbase: - node.site = HistorySiteRecord.by_loginbase(node.loginbase) - if node.site is None: + if agg.loginbase: + agg.site = HistorySiteRecord.by_loginbase(agg.loginbase) + if agg.site is None: # TODO: need a cleaner fix for this... - node.site = HistorySiteRecord.by_loginbase("pl") - if not node.site: - node.site = HistorySiteRecord.by_loginbase("ple") - + agg.site = HistorySiteRecord.by_loginbase("pl") + if not agg.site: + agg.site = HistorySiteRecord.by_loginbase("ple") - node.history = HistoryNodeRecord.by_hostname(node.hostname) + agg.history = HistoryNodeRecord.by_hostname(node.hostname) - node.ports = format_ports(node.port_status) + agg.ports = format_ports(node.port_status) try: exists = node.plc_node_stats['last_contact'] except: + # TODO: this should not assign to the fb object! node.plc_node_stats = {'last_contact' : None} - + + return agg class Root(controllers.RootController, MonitorXmlrpcServer): @@ -162,10 +171,10 @@ class Root(controllers.RootController, MonitorXmlrpcServer): def nodeview(self, hostname=None): nodequery=[] if hostname: - node = FindbadNodeRecord.get_latest_by(hostname=hostname) - # NOTE: reformat some fields. - prep_node_for_display(node) - nodequery += [node] + node = FindbadNodeRecord.get_latest_by(hostname=hostname) + # NOTE: reformat some fields. + agg = prep_node_for_display(node) + nodequery += [agg] return self.pcuview(None, None, hostname) # dict(nodequery=nodequery) @@ -181,41 +190,41 @@ class Root(controllers.RootController, MonitorXmlrpcServer): 'neverboot' : 0, 'pending' : 0, 'all' : 0, None : 0} for node in fbquery: # NOTE: reformat some fields. - prep_node_for_display(node) + agg = prep_node_for_display(node) #node.history.status #print node.hostname - if not node.history: + if not agg.history: continue - if node.history.status in ['down', 'offline']: + if agg.history.status in ['down', 'offline']: if node.plc_node_stats and node.plc_node_stats['last_contact'] != None: filtercount['down'] += 1 else: filtercount['neverboot'] += 1 - elif node.history.status in ['good', 'online']: + elif agg.history.status in ['good', 'online']: filtercount['boot'] += 1 - elif node.history.status in ['debug', 'monitordebug']: + elif agg.history.status in ['debug', 'monitordebug']: filtercount['debug'] += 1 else: - # TODO: need a better fix. filtercount - # doesn't maps to GetBootStates() on - # 4.3 so this one fails quite often. - if filtercount.has_key(node.history.status): - filtercount[node.history.status] += 1 + # TODO: need a better fix. filtercount + # doesn't maps to GetBootStates() on + # 4.3 so this one fails quite often. + if filtercount.has_key(agg.history.status): + filtercount[agg.history.status] += 1 # NOTE: apply filter if filter == "neverboot": if not node.plc_node_stats or node.plc_node_stats['last_contact'] == None: - query.append(node) + query.append(agg) elif filter == "all": - query.append(node) - elif filter == node.history.status: - query.append(node) + query.append(agg) + elif filter == agg.history.status: + query.append(agg) elif filter == 'boot': - query.append(node) + query.append(agg) widget = NodeWidget(template='monitorweb.templates.node_template') @@ -321,12 +330,12 @@ class Root(controllers.RootController, MonitorXmlrpcServer): pcus = {} for node in FindbadNodeRecord.query.filter_by(loginbase=loginbase): # NOTE: reformat some fields. - prep_node_for_display(node) - nodequery += [node] - if node.plc_pcuid: # not None - pcu = FindbadPCURecord.get_latest_by(plc_pcuid=node.plc_pcuid) - prep_pcu_for_display(pcu) - pcus[node.plc_pcuid] = pcu + agg = prep_node_for_display(node) + nodequery += [agg] + if agg.pcu: #.pcu.plc_pcuid: # not None + #pcu = FindbadPCURecord.get_latest_by(plc_pcuid=agg.plc_pcuid) + #prep_pcu_for_display(pcu) + pcus[agg.pcu.pcu.plc_pcuid] = agg.pcu for pcuid_key in pcus: pcuquery += [pcus[pcuid_key]] @@ -335,10 +344,10 @@ class Root(controllers.RootController, MonitorXmlrpcServer): print "pcuid: %s" % pcuid pcu = FindbadPCURecord.get_latest_by(plc_pcuid=pcuid) # NOTE: count filter - prep_pcu_for_display(pcu) - pcuquery += [pcu] + aggpcu = prep_pcu_for_display(pcu) + pcuquery += [aggpcu] if 'site_id' in pcu.plc_pcu_stats: - sitequery = [HistorySiteRecord.by_loginbase(pcu.loginbase)] + sitequery = [HistorySiteRecord.by_loginbase(aggpcu.loginbase)] if 'nodenames' in pcu.plc_pcu_stats: for nodename in pcu.plc_pcu_stats['nodenames']: @@ -347,19 +356,19 @@ class Root(controllers.RootController, MonitorXmlrpcServer): print "%s" % node.port_status print "%s" % node.to_dict() if node: - prep_node_for_display(node) - nodequery += [node] + agg = prep_node_for_display(node) + nodequery += [agg] if hostname and pcuid is None: node = FindbadNodeRecord.get_latest_by(hostname=hostname) # NOTE: reformat some fields. - prep_node_for_display(node) - sitequery = [node.site] - nodequery += [node] - if node.plc_pcuid: # not None - pcu = FindbadPCURecord.get_latest_by(plc_pcuid=node.plc_pcuid) - prep_pcu_for_display(pcu) - pcuquery += [pcu] + agg = prep_node_for_display(node) + sitequery = [agg.site] + nodequery += [agg] + if agg.plc_pcuid: # not None + #pcu = FindbadPCURecord.get_latest_by(plc_pcuid=node.plc_pcuid) + #prep_pcu_for_display(pcu) + pcuquery += [agg.pcu] return dict(sitequery=sitequery, pcuquery=pcuquery, nodequery=nodequery, actions=actions, exceptions=exceptions) @@ -429,22 +438,22 @@ class Root(controllers.RootController, MonitorXmlrpcServer): else: filtercount['pending'] += 1 - prep_pcu_for_display(node) + pcuagg = prep_pcu_for_display(node) # NOTE: apply filter if filter == "all": - query.append(node) + query.append(pcuagg) elif filter == "ok" and node.reboot_trial_status == str(0): - query.append(node) + query.append(pcuagg) elif filter == node.reboot_trial_status: - query.append(node) + query.append(pcuagg) elif filter == "pending": # TODO: look in message logs... if node.reboot_trial_status != str(0) and \ node.reboot_trial_status != 'NetDown' and \ node.reboot_trial_status != 'Not_Run': - query.append(node) + query.append(pcuagg) return dict(query=query, fc=filtercount) @@ -455,8 +464,8 @@ class Root(controllers.RootController, MonitorXmlrpcServer): nodequery = [] for node in FindbadNodeRecord.query.filter_by(loginbase=loginbase): # NOTE: reformat some fields. - prep_node_for_display(node) - nodequery += [node] + agg = prep_node_for_display(node) + nodequery += [agg] return dict(sitequery=sitequery, nodequery=nodequery, fc={}) @expose(template="monitorweb.templates.sitelist") diff --git a/web/MonitorWeb/monitorweb/templates/node_template.kid b/web/MonitorWeb/monitorweb/templates/node_template.kid index 795fdde..3e1583a 100644 --- a/web/MonitorWeb/monitorweb/templates/node_template.kid +++ b/web/MonitorWeb/monitorweb/templates/node_template.kid @@ -18,29 +18,29 @@ from links import * last_contact - node_id + node_id - your.host.org + your.host.org - ${node.loginbase} + ${node.node.loginbase} - ping - ssh - boot + ping + ssh + boot - - + + diff --git a/web/MonitorWeb/monitorweb/templates/nodelist.kid b/web/MonitorWeb/monitorweb/templates/nodelist.kid index 2c3cd00..29a09fe 100644 --- a/web/MonitorWeb/monitorweb/templates/nodelist.kid +++ b/web/MonitorWeb/monitorweb/templates/nodelist.kid @@ -17,8 +17,8 @@ from links import *
- | -
+ | +
diff --git a/web/MonitorWeb/monitorweb/templates/pculist.kid b/web/MonitorWeb/monitorweb/templates/pculist.kid index 6671c70..2d06887 100644 --- a/web/MonitorWeb/monitorweb/templates/pculist.kid +++ b/web/MonitorWeb/monitorweb/templates/pculist.kid @@ -58,14 +58,14 @@ from links import * @@ -76,15 +76,15 @@ from links import * - - + +
${node.loginbase} - +
- ${pcu_name(node.plc_pcu_stats)} - + ${pcu_name(node.pcu.plc_pcu_stats)} +
diff --git a/web/MonitorWeb/monitorweb/templates/pcuview.kid b/web/MonitorWeb/monitorweb/templates/pcuview.kid index 6e241d5..b3aa16d 100644 --- a/web/MonitorWeb/monitorweb/templates/pcuview.kid +++ b/web/MonitorWeb/monitorweb/templates/pcuview.kid @@ -55,23 +55,23 @@ from links import * - + - history + history - - ${pcu_name(pcu.plc_pcu_stats)} + + ${pcu_name(agg.pcu.plc_pcu_stats)} - - + + - 80 - 

-					
-					
+					

+					
+					
 				
 			
 		
@@ -154,31 +154,31 @@ from links import *
 				
 			
 			
-				
+				
 					
-					history
-					
-						
-							${node.hostname}
+					history
+					
+						
+							${agg.node.hostname}
 					
-					
-					
+					
+					
 					
-						80
 					
 					
 						
-	  					
 					
 					
 						
-	  					
@@ -186,7 +186,7 @@ from links import *
 					
 						
 	  					
@@ -231,7 +231,11 @@ from links import *
 					
 					
 							${act.message_id}
-					

+					
+						latest bm log
+					
+					
+						

 				
 			
 		
-- 
2.43.0