Over 8Tbyte DiskSpace cant get correct TotalSpace

[katayama]

I use large Disk Space (Over 10Tbyte) in Windows.
I graphed disk usage in cacti 0.8.7b but Total Disk Space cant get correct data from snmp.(attached graph)

I check snmpwalk. I get following data.

HOST-RESOURCES-MIB::hrStorageDescr.1 = STRING: C:\ Label: Serial Number aaaaaaaa
HOST-RESOURCES-MIB::hrStorageDescr.2 = STRING: D:\
HOST-RESOURCES-MIB::hrStorageDescr.3 = STRING: E:\ Label: Serial Number ccccccccc
HOST-RESOURCES-MIB::hrStorageDescr.4 = STRING: F:\ Label: Serial Number bbbbbbbb
HOST-RESOURCES-MIB::hrStorageAllocationUnits.1 = INTEGER: 4096 Bytes
HOST-RESOURCES-MIB::hrStorageAllocationUnits.2 = INTEGER: 0 Bytes
HOST-RESOURCES-MIB::hrStorageAllocationUnits.3 = INTEGER: 4096 Bytes
HOST-RESOURCES-MIB::hrStorageAllocationUnits.4 = INTEGER: 4096 Bytes
HOST-RESOURCES-MIB::hrStorageSize.1 = INTEGER: 7865817
HOST-RESOURCES-MIB::hrStorageSize.2 = INTEGER: 0
HOST-RESOURCES-MIB::hrStorageSize.3 = INTEGER: 114190019
HOST-RESOURCES-MIB::hrStorageSize.4 = INTEGER: -876942081

I guess scripts/ss_host_disk.php need fix.
I search this forum and bug report site , but I cant find patch relate this things.

anyone fixed it?

[katayama]

sorry , I fixed myself.

patch is here

Code: Select all

--- ss_host_disk.php.back       2008-03-28 18:32:32.000000000 +0900
+++ ss_host_disk.php    2008-08-20 12:01:07.000000000 +0900
@@ -75,7 +75,12 @@
 
                if (($arg == "total") || ($arg == "used")) {
                        $sau = eregi_replace("[^0-9]", "", db_fetch_cell("select field_value from host_snmp_cache where host_id=$host_id and field_name='hrStorageAllocationUnits' and snmp_index='$index'"));
-                       return cacti_snmp_get($hostname, $snmp_community, $oids[$arg] . ".$index", $snmp_version, $snmp_auth_username, $snmp_auth_password, $snmp_auth_protocol,$snmp_priv_passphrase,$snmp_priv_protocol, $snmp_context, $snmp_port, $snmp_timeout, read_config_option("snmp_retries"), SNMP_POLLER)* $sau;
+                       $snmp_data = cacti_snmp_get($hostname, $snmp_community, $oids[$arg] . ".$index", $snmp_version, $snmp_auth_username, $snmp_auth_password, $snmp_auth_protocol,$snmp_priv_passphrase,$snmp_priv_protocol, $snmp_context, $snmp_port, $snmp_timeout, read_config_option("snmp_retries"), SNMP_POLLER);
+                       if( preg_match('/^-/',$snmp_data) ){
+                               return (abs($snmp_data) + 2147483647) * $sau;
+                       } else {
+                               return $snmp_data * $sau;
+                       }
                }else{
                        return cacti_snmp_get($hostname, $snmp_community, $oids[$arg] . ".$index", $snmp_version, $snmp_auth_username, $snmp_auth_password, $snmp_auth_protocol,$snmp_priv_passphrase,$snmp_priv_protocol, $snmp_context, $snmp_port, $snmp_timeout, read_config_option("snmp_retries"), SNMP_POLLER);
                }

and fixed graph is attached.

[gandalf]

Thank you for the solution. Please post a bug as given by http://www.cacti.net/bugs.php and attach all information. Then we'll integrate it into cacti asap
Reinhard

[nileshmistry73]

Hello

I am running the same version of cacti and applied the patch manually, but still having the same problem.

Is there anything I am missing other then applying the changes to ss_host_disk.php?

Nilesh

[BSOD2600]

Is your windows snmp counter also rolling over, like the OP was? More of an issue with the SNMP agent instead of Cacti not reporting the correct value, IMO.

[opoplawski]

Code: Select all

                       if( preg_match('/^-/',$snmp_data) ){
                               return (abs($snmp_data) + 2147483647) * $sau;

Integer wrapping doesn't work that way. I think this is what you want:

Code: Select all

                        if($snmp_data<0){
                                return ($snmp_data + 4294967296) * $sau;

[kevinadmin]

ths i solved my problem.

but the disk used percent can not display,and it's no data . why?

i found the same in the smaller disk also.

it is the bug on 0.8.7e version?

[kevinadmin]

Can not get the data
and in the ss_host_disk.php script, can not find about percent any more

[mbails]

Can anybody provide a patched ss_host_disk.php?

[computer_guru]

The fix mentioned in this thread for large hard drives appears to have never been integrated into the cacti code. Any reason for this?

I checked the changes and I did not see this addition:
http://svn.cacti.net/viewvc/cacti/branc ... p?view=log


Anybody find a good solution?

[gandalf]

Any reason for this?

Lack of memory and no bug report. Sorry, sometimes it's that simple. But it's hard to track forum messages AND bug reports. So we prefer doing the last.
R.

[computer_guru]

I am currently trying to perfect the PHP code. Applying the fix above put the free space in place of the total space.

Shall I put in a bug report?

[computer_guru]

Attached is my version of the ss_host_disk.php file. It should be able to handle 16 TB drives and smaller with a ntfs cluster size of 4096.

If you want to handle larger drives, than use a larger NTFS cluster size.

More about ntfs cluster size:
http://www.ntfs.com/ntfs_optimization.htm


[update]
File removed. Download the newer version on page two of this thread.

[gandalf]

I am currently trying to perfect the PHP code. Applying the fix above put the free space in place of the total space.

Shall I put in a bug report?

Yes, please. See http://www.cacti.net/bugs.php
R.

BTB: Do you know this one? http://docs.cacti.net/usertemplate:data ... disk_usage
Pure SNMP, fast, does take block size into account automatically. But not tested against such a big device as I do not own such

[computer_guru]

If found this bug as already reported:
http://bugs.cacti.net/view.php?id=1292


I updated it with my proposed solution.