Description of problem:

A directory server instance fails to start up with the following error.

----
[04/Sep/2017:20:18:14.530818672 +0900] - ERR - log_set_logsize - Invalid value
for Maximum log size:Maxlogsize:300 (MB) exceeds Maxdisksize:100 (MB)
[04/Sep/2017:20:18:14.535062585 +0900] - ERR - dse_read_one_file - The entry
cn=config in file /etc/dirsrv/slapd-ds2/dse.ldif (lineno: 10) is invalid, error
code 1 (Operations error) -
[04/Sep/2017:20:18:14.569222415 +0900] - ERR - init_dse_file - Could not load
config file [dse.ldif]
----

As far as I confirmed, this happens with the following conditions a, b and c.

a. package version is 389-ds-base-1.3.6.1-16.el7.
b. value of parameters.

  nsslapd-*log-maxlogsize is greater than 100 (MB).

c. The order of parameters in cn=config .

- The problem happens.
---
dn: cn=config
(snip)
nsslapd-errorlog-maxlogsize: 300
nsslapd-errorlog-logmaxdiskspace: 500
(snip)
---

- The problem does not happen.
---
dn: cn=config
(snip)
nsslapd-errorlog-logmaxdiskspace: 500
nsslapd-errorlog-maxlogsize: 300
(snip)
---

Version-Release number of selected component (if applicable):
389-ds-base-1.3.6.1-16.el7.x86_64
Red Hat Directory Server 10

How reproducible:
Always

Steps to Reproduce:
1) Prepare a Directory Server Instance with 389-ds-base-1.3.6.1-16.el7. (RHDS
10)
2) Stop the directory server instance.
3) Add the following two lines to cn=config entry in dse.ldif.

---
dn: cn=config
(snip)
nsslapd-errorlog-maxlogsize: 300
nsslapd-errorlog-logmaxdiskspace: 500
(snip)
----
4) Start the instance but it fails.

[04/Sep/2017:20:18:14.530818672 +0900] - ERR - log_set_logsize - Invalid value
for Maximum log size:Maxlogsize:300 (MB) exceeds Maxdisksize:100 (MB)
[04/Sep/2017:20:18:14.535062585 +0900] - ERR - dse_read_one_file - The entry
cn=config in file /etc/dirsrv/slapd-ds2/dse.ldif (lineno: 10) is invalid, error
code 1 (Operations error) -
[04/Sep/2017:20:18:14.569222415 +0900] - ERR - init_dse_file - Could not load
config file [dse.ldif]

5) Edit the dse.ldif file again and change the order of attributes.

---
dn: cn=config
(snip)
nsslapd-errorlog-logmaxdiskspace: 500
nsslapd-errorlog-maxlogsize: 300
(snip)
----

6) Start the instance again. It can start up without errors.

[04/Sep/2017:20:19:51.699856551 +0900] - INFO - main - 389-Directory/1.3.6.1
B2017.158.1159 starting up
[04/Sep/2017:20:19:51.712940094 +0900] - INFO -
ldbm_instance_config_cachememsize_set - force a minimal value 512000
[04/Sep/2017:20:19:51.721326096 +0900] - NOTICE - ldbm_back_start - found
1015004k physical memory
[04/Sep/2017:20:19:51.723214606 +0900] - NOTICE - ldbm_back_start - found
703204k available
[04/Sep/2017:20:19:51.725416288 +0900] - NOTICE - ldbm_back_start - cache
autosizing: db cache: 40600k
[04/Sep/2017:20:19:51.727704792 +0900] - NOTICE - ldbm_back_start - cache
autosizing: userRoot entry cache (1 total): 65536k
[04/Sep/2017:20:19:51.732686002 +0900] - NOTICE - ldbm_back_start - total cache
size: 110854274 B;
[04/Sep/2017:20:19:51.848819439 +0900] - INFO - slapd_daemon - slapd started.
Listening on All Interfaces port 12389 for LDAP requests

Actual results:
DS instance fails to start.

Expected results:
DS instance did not to start.

Additional info:
The phenomenon doesn't occur with old 389-ds-base package like
389-ds-base-1.3.5.10-18.el7_3.
This problem was reported by a customer who updated the 389-ds-base to
389-ds-base-1.3.6.1-16.el7.