IBM managed system IP change caused duplicate data directories
Hello,
We are experiencing an issue with LPAR2RRD after an IP address change on one of our managed systems, and I would appreciate guidance on the best approach to resolve this.
**Environment:**
- LPAR2RRD version: 7.95
- Platform: IBM Power Systems (AIX)
**Problem Description:**
Until November 22nd, 2024, performance data for LPARs hosted on physical machine "S0-P09-15" was being collected correctly. Since November 22nd, we now see two separate entries in the LPAR2RRD web interface under "IBM Power Systems":
- A physical machine named "10.0.0.15" (the old IP address)
- A physical machine named "S0-P09-15" (the current hostname)
These both refer to the same physical server. The issue appears to have been triggered by a change in how the system is identified.
**Current Situation:**
On the LPAR2RRD server, we have two separate data directory trees:
[root@infra-xorux data]# pwd
/home/lpar2rrd/lpar2rrd/data
[root@infra-xorux data]# ls -ld *10.0.0.15* *S0-P09-15*
drwxr-xr-x 5 lpar2rrd lpar2rrd 59 Dec 15 2022 10.0.0.15
drwxr-xr-x 3 lpar2rrd lpar2rrd 24 Nov 22 15:00 S0-P09-15
- Historical data (up to Nov 22nd) is stored in `./10.0.0.15/`
- New data (since Nov 22nd) is being stored in `./S0-P09-15/`
Example of duplicate RRD files for the same LPAR:
# Historical data (last updated Nov 22)
./10.0.0.15/10.16.0.110/aix72-test1.rrm (3825840 bytes, updated Nov 22 08:20)
./10.0.0.15/10.16.0.110/aix72-test1.rsm (957048 bytes, updated Nov 22 08:20)
./10.0.0.15/10.16.0.110/aix72-test1.rvm (58072 bytes, updated Nov 22 08:20)
./10.0.0.15/10.16.0.110/aix72-test1.grm (12432216 bytes, updated Nov 22 08:20)
# Current data (being updated)
./S0-P09-15/10.16.0.110/aix72-test1.rrm (3825840 bytes, updated Nov 28 17:20)
./S0-P09-15/10.16.0.110/aix72-test1.rsm (957048 bytes, updated Nov 28 17:20)
./S0-P09-15/10.16.0.110/aix72-test1.rvm (58072 bytes, updated Nov 28 17:20)
./S0-P09-15/10.16.0.110/aix72-test1.grm (12432216 bytes, updated Nov 28 17:20)
**Questions:**
1. Is this a known issue? Is there any official documentation about handling physical server identification changes?
2. What is the recommended procedure to merge these two data trees while preserving all historical data?
- Should I manually merge the RRD files using rrdtool dump/restore?
- Is there a built-in script or utility for this purpose ?
3. How can I prevent this from happening in the future? Should I configure aliases in `etc/alias.cfg`?
Any guidance on the recommended approach would be greatly appreciated. I want to ensure we preserve our historical performance data while maintaining data integrity.
Thank you for your assistance!
Best regards.
Comments
-
The issue occurred in **November 2025** (not 2024).
All dates mentioned should be read as 2025.
- Issue started: November 22nd, **2025**
- Current date: November 28th, 2025
Apologies for the typo.
Best regards.
-
Hi,
ok, it might happen that when server on the HMC is in some weird status, we have seen that a few time in past years.
Data cannot be merged, if you want to keep history you have to remove actual S0-P09-15 (which has just 2 weeks data only) and rename 10.0.0.15 to S0-P09-15.
There will be 2 weeks gap in data.
Make a backup before start with anything.
2.data cannot be merged
3.no real prevention, we are not sure how this can happen in the HMC, no way to prevent it on our side, basically HMC reporting it
real prevention is migrating to XorMon, it cannot happen there due to its implementation there.
---
Download XorMon from https://xormon.com/download.php and install it according to the instructions at https://xormon.com/install-ng.php.
The major news compared to older tools is available at https://xormon.com/documentation-ng.php.
It comes with a two-month initial trial with no restrictions.
Use the trial to test as many technologies as possible. As soon as you apply the license, any unlicensed technologies will switch to the free edition with some restrictions.
Ask us for a license whenever you need one, and we will provide it.
Note that LPAR2RRD, STOR2RRD, and XorMon Original are no longer being developed beyond bug fixes and minor enhancements.
Read more about data migration: https://xormon.com/data-migration.php
--
Howdy, Stranger!
Categories
- 1.7K All Categories
- 113 XorMon
- 26 XorMon Original
- 169 LPAR2RRD
- 14 VMware
- 19 IBM i
- 2 oVirt / RHV
- 4 MS Windows and Hyper-V
- Solaris / OracleVM
- 1 XenServer / Citrix
- Nutanix
- 8 Database
- 2 Cloud
- 10 Kubernetes / OpenShift / Docker
- 139 STOR2RRD
- 20 SAN
- 7 LAN
- 19 IBM
- 7 EMC
- 12 Hitachi
- 5 NetApp
- 17 HPE
- 1 Lenovo
- 1 Huawei
- 3 Dell
- Fujitsu
- 2 DataCore
- INFINIDAT
- 4 Pure Storage
- Oracle