Manage Plugin 0.6.2 for monitoring networks, servers...

General discussion about Plugins for Cacti

Moderators: Moderators, Developers

Post Reply
Author
Message
jherrinike
Posts: 22
Joined: Wed May 10, 2006 2:46 pm
Location: Memphis, TN USA

#61 Post by jherrinike » Wed Jun 14, 2006 5:54 pm

The patched files fixed the two errors I was seeing. I am also seeing what I think someone else has reported about the uptime being shown twice on a device. One of my switches has this next to it for uptime. 150d 6h 23m150d 6h 23m I have even tried deleting the device out and recreating it but it still comes back with the host.

Two other things I have noticed is that the uptime is not right for my switches that are running CatOS instead of IOS and even though several of the devices are identical in the manage screen window some of them have the telnet service listed next to the green box and some do not even though the settings for both are identical. The other issue seems to also involve the CatOS operating system. The manage plugin will sometimes show the device as down when it is not. The monitor plugin at the same time stills shows the device is up.

This beta plugin is coming along well. I can not wait to have a full production one going.

User avatar
gilles
Cacti User
Posts: 250
Joined: Tue Jul 12, 2005 8:16 pm

#62 Post by gilles » Wed Jun 14, 2006 6:04 pm

jherrinike wrote:One of my switches has this next to it for uptime. 150d 6h 23m150d 6h 23m I have even tried deleting the device out and recreating it but it still comes back with the host.
in the tab manage, can you tell me how many times this switch is listed ? 1 or 2 ?

jherrinike
Posts: 22
Joined: Wed May 10, 2006 2:46 pm
Location: Memphis, TN USA

#63 Post by jherrinike » Wed Jun 14, 2006 6:22 pm

There is 2 listings of this device on the manage screen.

User avatar
gilles
Cacti User
Posts: 250
Joined: Tue Jul 12, 2005 8:16 pm

#64 Post by gilles » Wed Jun 14, 2006 6:35 pm

i don't know why this happend...
do you remember precisly what you have done ?

to delete the "bad" switch, with phpmyadmin :
in table manage_host, index id, all id's must be unique
list them by id and delete all which are not unique

jherrinike
Posts: 22
Joined: Wed May 10, 2006 2:46 pm
Location: Memphis, TN USA

#65 Post by jherrinike » Wed Jun 14, 2006 6:50 pm

I know it happened when I was changing the type and the adding in the device template. Not sure what went wrong with it. Maybe I hit the back arrow intead of selecting save.

I have not messed with phpadmin so I am going to have to work on that.

User avatar
gilles
Cacti User
Posts: 250
Joined: Tue Jul 12, 2005 8:16 pm

#66 Post by gilles » Wed Jun 14, 2006 7:14 pm

i don't know if it's that but, yes, in a certain circonstance, the back button has inserted a second time my host

so this is the patch for this problem

IT IS INCLUDED IN 0.3.2

jherrinike, if you don't want to use phpmyadmin, just choose "don't manage" for this host
verify it is no more listed in tab manage
then click "manage" for this host and reconfigure it
Attachments
listmanage2.php.txt
listmanage2.php
(30.37 KiB) Downloaded 652 times

User avatar
gilles
Cacti User
Posts: 250
Joined: Tue Jul 12, 2005 8:16 pm

#67 Post by gilles » Wed Jun 14, 2006 8:23 pm

OK
since version 0.1, i think now i have the minimum (device types, templates, groups) !

next version will take approximatively 3 weeks to come

this is my small roadmap (it can change) :

0.4 :
- will alert by "net send" one or more Windows boxes (this is pretty urgent for me, and i don't want to wait for local syslogging)
- will include Windows Services check by SNMP
- will add List view (like http://demo.bb4.com/bb/bb.html)

0.5 :
- possibility to select email-alert of all 3 events: down/up/reboot
- will add refresh on tab Manage
- will let you sort hosts by hostname, description, type etc... in tab manage
- will let you choose to list or not last events in Tab Manage

0.6 :
- will add local syslogging
- possibility to play a sound if an alert occurs
- will add UDP check
- export alerts to .csv in Event Reporting

0.7 :
- will add unique email addresses for each hosts
- mouse-over effect in Simple View
- make it independant of Thold plugin

torstentfk
Cacti User
Posts: 367
Joined: Tue Apr 05, 2005 9:52 am
Location: Munich, Germany

#68 Post by torstentfk » Thu Jun 15, 2006 6:28 am

Hi,
gilles wrote:
torstentfk wrote: - Template selections in device view are not saved
Torsten
if i save the template, i will need to write code to prevent you to delete a used template, etc...so i choose to not save it
so it is not yet implemented?

torstentfk wrote: - how do you guess the hosttype? Cisco switch (catos) is guested as winnt???
Torsten
i use index "id" in cacti table "host_template"
for all of us, 5 is "Cisco Router", 7 is "Windows 2000/XP Host", 8 is "Local Linux Machine" etc
the id's after 8 depends of what templates you have imported
for me, the id 9 is "Windows NT Host"
for you you, it will be different, this is why is use the term "trying"
where can I change this in the cacti code?

torstentfk wrote: -correct guest hosts (cisco router) are not displayed correct: the uptime is not updated (N/A), after changing the host device type from router to router the device works correct
Torsten
from router to router ?
i'm don't have this bug
from router type to router type - you are right. I think this problem of displaying the uptime is solved if I go to a soec. device and klick save (regardless of changing something)
torstentfk wrote: - the mouse-over effect of the monitor-plugin
Torsten
i don't know what to display (and i don't want to deal with multiple browsers)
ok - stay independed from browsers...
torstentfk wrote: - getting the list of last reboots in the tabed view instead of the console view
(View reboots and then the symbols)
Torsten
where ? in another tab ? on mouse over ?
not in another tab (or why not) I think of a page splitted into two parts: first the list of the rebots (as seen in the uptime plugin) and then the host list.
torstentfk wrote: - display of the last 10 events and not haveing to select a specific device
(event view not in the console)
Torsten
why not
because I must give users the console right to dispay it
torstentfk wrote: - new setting view as used by thold (not subsection under MISC)
Torsten
others plugins use MISC, so i use MISC too (btw, it's right THOLD have is own tab but this is because it have much options)
how many option are you planning for future?

[/quote]

r_chetanjain
Cacti User
Posts: 83
Joined: Fri Aug 22, 2003 4:33 am

mysql version 5

#69 Post by r_chetanjain » Fri Jun 16, 2006 4:41 am

Hi giles...I think there is a prob with the mysql version 5, as i am able to run manage 0.3 on my test box which is running mysql 4.1 but its not able to create the tables in mysql 5.0 (Fedora core 5 )...

When i login to cacti..it says cacti.manage_host table not found...can u post mysql 5 compatible .sql file

My cacti.log show some error like this

06/16/2006 05:42:01 AM - MANAGE: Poller[0] Checking Manage
06/16/2006 05:42:01 AM - CMDPHP: Poller[0] ERROR: SQL Assoc Failed "SELECT host.id, host.hostname, host.description, host.snmp_community, host.snmp_version, host.snmp_username, host.snmp_password, host.snmp_port, host.snmp_timeout, manage_host.uptime, manage_host.statut, manage_host.force FROM host,manage_host where manage = 'on' and disabled <> 'on' and host.id=manage_host.id"

Can u give a step by step installation procedure......


Chetan

cotontige
Posts: 6
Joined: Tue Apr 25, 2006 7:14 am

thanks

#70 Post by cotontige » Fri Jun 23, 2006 7:53 am

great plugins !!!! 8)

jherrinike
Posts: 22
Joined: Wed May 10, 2006 2:46 pm
Location: Memphis, TN USA

#71 Post by jherrinike » Fri Jun 23, 2006 9:53 am

This plugin is running okay on my test cacti system with mysql 5.

mories
Posts: 8
Joined: Fri Dec 30, 2005 1:45 pm

#72 Post by mories » Sun Jun 25, 2006 3:07 pm

Hello gilles,

I think I have found a bug in viewalerts.pnp and solved it ;-)

When navigating through the 'Event reporting' page using the numbered index in the header I had the following error.

http://cacti/plugins/manage/viewalerts.php&page=2
404 The requested URL /plugins/manage/viewalerts.php&page=2 was not found on this server.
That is correct because php expects an ? instead of an &.

So I was so free to modify viewalerts.php and now the problem is solved, it now remembers which options you choose in the filter when navigating with '<< Previous', 'numbers' or 'Next >>'

Below you find the changes i made.

Mories

changed line 267
from : $url_page_select = get_page_list($_REQUEST["page"], MAX_DISPLAY_PAGES, read_config_option("num_rows_device"), $total_rows, "viewalerts.php");
to : $url_page_select = get_page_list($_REQUEST["page"], MAX_DISPLAY_PAGES, read_config_option("num_rows_device"), $total_rows, "viewalerts.php?host_managed=".$_REQUEST["host_managed"]."&event=".$_REQUEST["event"]);

changed line 278
from : <strong><< "; if ($_REQUEST["page"] > 1) { $nav .= "<a class='linkOverDark' href='viewalerts.php?" . "page=" . ($_REQUEST["page"]-1) . "'>"; } $nav .= "Previous"; if ($_REQUEST["page"] > 1) { $nav .= "</a>"; } $nav .= "</strong>
to : <strong><< "; if ($_REQUEST["page"] > 1) { $nav .= "<a class='linkOverDark' href='viewalerts.php?" . "page=" . ($_REQUEST["page"]-1) . "&host_managed=".$_REQUEST["host_managed"]."&event=".$_REQUEST["event"] . "'>"; } $nav .= "Previous"; if ($_REQUEST["page"] > 1) { $nav .= "</a>"; } $nav .= "</strong>

changed line 284
from : <strong>"; if (($_REQUEST["page"] * read_config_option("num_rows_device")) < $total_rows) { $nav .= "<a class='linkOverDark' href='viewalerts.php?" . "page=" . ($_REQUEST["page"]+1) . "'>"; } $nav .= "Next"; if (($_REQUEST["page"] * read_config_option("num_rows_device")) < $total_rows) { $nav .= "</a>"; } $nav .= " >></strong>
to : <strong>"; if (($_REQUEST["page"] * read_config_option("num_rows_device")) < $total_rows) { $nav .= "<a class='linkOverDark' href='viewalerts.php?" . "page=" . ($_REQUEST["page"]+1) . "&host_managed=".$_REQUEST["host_managed"]."&event=".$_REQUEST["event"] . "'>"; } $nav .= "Next"; if (($_REQUEST["page"] * read_config_option("num_rows_device")) < $total_rows) { $nav .= "</a>"; } $nav .= " >></strong>

User avatar
NHDave
Cacti User
Posts: 61
Joined: Sun Mar 20, 2005 9:14 am
Contact:

#73 Post by NHDave » Mon Jun 26, 2006 8:52 am

Is there a way to change the sort order on the "mange" tab. Seems like they are displaying in the order they have been added rather then alphabetically.

Great plugin!

Dave

torstentfk
Cacti User
Posts: 367
Joined: Tue Apr 05, 2005 9:52 am
Location: Munich, Germany

#74 Post by torstentfk » Tue Jun 27, 2006 2:18 am

Hi,

some point to mention:
- it would be nice to have the actual time of the reboot/host down event in the email body because receiving email could take some time.
- does the plugin respect the setting of the thold plugin that the device must be down XXX cylces before alerting?
- possibility to enter a filter at the device section in the console-"Device-Managing" (there is space in the first line right of the device-selection) as found at the 'normal' device selection
- at event-reporting:display only devices that have a alarm;
- the selection of "type of event" does not work

Thanks for this plugin - it is installed in a productiv enviroment in my company!

Torsten

User avatar
NHDave
Cacti User
Posts: 61
Joined: Sun Mar 20, 2005 9:14 am
Contact:

#75 Post by NHDave » Tue Jun 27, 2006 7:15 am

NHDave wrote:Is there a way to change the sort order on the "mange" tab. Seems like they are displaying in the order they have been added rather then alphabetically.

Great plugin!

Dave
Found it, please ignore.. :wink:

Post Reply