Create/manager parameter groupt

With parameter groups, you can view and manage all the parameters required for Redis OSS settings in MemStore according to the engine version.
You can also manage the parameters of clusters connected to that parameter group in bulk.

The parameter list and underlying concepts provided by MemStore are compatible with Redis OSS. Since changing parameters can affect the MemStore service, it is recommended to check the appropriate parameters in advance.

Here’s how to manage parameter groups in the MemStore service.

info

• Events such as creation and modification of parameter groups can be received through Alert Center > Event notifications.
• For configurable event types, refer to Cloud Trail > Key concepts, and for alert policy settings, refer to Alert Center > Create and manage alert policies.

View parameter group list

You can view the list of default and custom parameter groups provided by MemStore.

1. Go to the KakaoCloud Console > Data Store > MemStore.

2. Click the Parameter groups menu, then select either the Custom groups or Default groups tab.

3. Navigate to the parameter group list corresponding to the selected tab.

   Item	Description
   Name	Name of the parameter group
   Cluster mode	Whether cluster mode is enabled
   Family	Engine version family compatible with the group
   Connected cluster count	Number of clusters linked to this group - If any cluster has an abnormal parameter application state, a warning icon is displayed
   Description	Description of the parameter group
   [⋮] button	Available only for custom groups - Modify: Edit parameters - Copy: Create a new group using the same parameters - Reset to default: Reset group parameters to initial values at creation - Rollback: Revert group parameters to previous values (not available if no previous modifications or rollback already used) - Delete: Delete the parameter group

View parameter group details

You can view detailed information of a MemStore parameter group.

1. Go to the KakaoCloud Console > Data Store > MemStore.

2. Click Parameter groups, then select either Custom groups or Default groups tab according to your classification.

3. Navigate to the list and select the parameter group you want to view.

   Parameters tab

   Item	Description
   Name	Parameter name
   Value	Parameter value
   Type	Parameter type – Some Dynamic parameters apply globally and may require session refresh
   Data type	Data type of the parameter
   Editable	Not editable, Editable, Edited - Only Editable parameters can be modified; if modified, status becomes Edited

   Connected clusters tab

   Item	Description
   Name	Cluster name
   Status	Cluster status
   Parameter apply status	Status of parameter application; for details see Parameter application status
   Engine version	Redis OSS engine version of the cluster
   Node type	Instance type used for the cluster
   Cluster mode	Whether cluster mode is used (Enabled, Disabled)
   High availability	Whether high availability is used (Enabled, Disabled)
   [Retry] button	Appears when apply status is Error‑Sync or Suspended, allows re-applying parameters to the cluster

   Recent events tab

   You can view major events of the parameter group from the last 7 days.

   Item	Description
   Time	Time when the event occurred
   Event name	Name of the event
   Details	Description of event result, cluster name, etc.

   • Parameter group event types

   Group type	Event name	Description
   Common	Apply Parameter Group	Triggered when parameter group application is requested
   	Apply Parameter Group Failed	Triggered when applying to clusters failed
   	Apply Parameter Group Complete	Triggered when all connected clusters have completed application
   	Apply Parameter Group To Cluster	Triggered when applying to a specific cluster begins
   	Apply Parameter Group To Cluster Failed	Triggered when applying to a specific cluster fails
   	Apply Parameter Group To Cluster Complete	Triggered when a specific cluster completes application
   Custom	Create Parameter Group	Triggered when parameter group creation is requested
   	Update Parameter Group	Triggered when group modification is requested
   	Update Parameters	Triggered when parameter values are modified
   	Reset Parameters	Triggered when initialization is requested
   	Rollback Parameters	Triggered when rollback is requested

Create parameter group

1. Go to the KakaoCloud Console > Data Store > MemStore.

2. Click Parameter groups, then select the Custom groups tab.

3. On the Custom groups tab, click the icon > Create parameter group.

4. On the Create parameter group page, fill in the necessary fields. Parameter values can be edited after group creation.

   Item	Category	Description
   Parameter group name	Base setting	Name of the group - Duplicate names are not allowed within the same project
   Description (optional)		Up to 100 characters
   Family	Parameter group setting	MemStore engine version family
   Cluster mode		Whether cluster mode is enabled
   Base parameter group		Default parameter group offered for the selected engine version
   Parameters		Initial parameter values for the group

Modify parameter group

For custom parameter groups, you can modify description or change parameter values after creation. Only parameters marked Editable can be modified. Modified parameters will show Edited status.

info

• Some Dynamic parameters apply globally so a session refresh may be required.
• Parameters cannot be modified if connected clusters have apply status Pending or Applying.
• On the modification page, checkboxes remain selected while scrolling through editable parameters.

caution

• Some parameter changes may affect MemStore service operations. For details, see Parameters to be careful when modifying.
• If issues occur after modification, it is recommended to use group Copy, Retry/Rollback/Reset to secure the previous settings.

1. Go to the KakaoCloud Console > Data Store > MemStore.

2. Click Parameter groups, then select the Custom groups tab.

3. Select the group to modify and click [⋮] > Modify.

4. On the Modify parameter group page, update fields as needed.

   Item	Category	Description
   Parameter group name	Base info	Name of the group
   Description (optional)		Up to 100 characters
   Editable parameters		List of parameters that can be changed - Modified parameters show status Edited

5. If multiple clusters are linked to the parameter group, choose application method:
   - Apply in batch: Apply modified parameters to all clusters at once.
   - Apply sequentially: Apply changes to clusters one after another.

Parameters to be careful when modifying

When modifying parameter values in a custom parameter group, some parameters may affect MemStore service operations. Please refer to the details below.

info

repl-backlog

• A buffer on the replica node holding data sent from the primary node when replication is interrupted by network disconnection.
  
• Upon disconnection, the primary accumulates sync data in the backlog, and attempts partial sync (psync) when the replica reconnects.
  
• The primary stores backlog data only up to a certain time (repl‑backlog‑ttl). After this, the backlog buffer is released. If released and replica reconnects, a full sync is required.
  
• Related parameters: repl‑backlog‑size, repl‑backlog‑ttl

client‑output‑buffer‑limit‑normal‑hard‑limit

Type	Range	Description
integer	0~	The maximum memory usage of client output buffers. Set to 0 for no limit. Potential issue If buffer memory reaches this value, client connection will be terminated. Setting a too low value may prevent service provisioning. It is recommended to set a sufficient number or leave default (0).

client‑output‑buffer‑limit‑normal‑soft‑limit

Type	Range	Description
integer	0~	Threshold for client output buffer memory usage, allowed for a period of client‑output‑buffer‑limit‑normal‑soft‑seconds. Set to 0 for no threshold. Potential issue If buffer memory exceeds this value for longer than the soft‑seconds parameter, client connection will be terminated. A too low value may prevent service provisioning. It is recommended to set a sufficient number or leave default (0).

client‑output‑buffer‑limit‑normal‑soft‑seconds

Type	Range	Description
integer	0~2147483647	Time (in seconds) a client can exceed the soft‑limit threshold before being disconnected. Set to 0 for no time limit. Potential issue If buffer exceeds the soft‑limit for longer than this value, connection will be terminated. Setting a too low value may prevent service provisioning. It is recommended to set a sufficient number or leave default (0).

min‑replicas‑max‑lag

Type	Range	Description
integer	0~2147483647	Maximum lag (in pings) that the primary node allows for any replica node. Potential issue If a replica fails to respond within this parameter, it will be considered unusable until reconnection. For stability, it is recommended to set an appropriate value.

min‑replicas‑to‑write

Type	Range	Description
integer	0~2147483647	Minimum number of replica nodes that must be present for primary node to accept write requests. Potential issue If the number of replicas does not reach this count, the primary will refuse writes. Example: If min‑replicas‑to‑write: 5 but cluster has only 2 replica nodes: primary.memstore:6379> SET hello memstore(error) NOREPLICAS Not enough good replicas to write.

repl‑backlog‑size

Type	Range	Description
integer	16384~	Size of the backlog buffer in the primary node for data sync to replicas. Potential issue If replication lag exceeds the backlog size, full sync occurs, which may degrade service performance or availability. For optimal operation, set a sufficient value based on workload.

repl‑backlog‑ttl

Type	Range	Description
integer	0~2147483647	Time period (in seconds) the primary node retains backlog data for replica sync. Set to 0 to never release backlog. Potential issue If backlog is released before replica reconnects, full sync occurs. Too small a value may lead to frequent full syncs, impacting performance and availability. For workload, set a sufficient value.

maxclients

Type	Range	Description
integer	100~65000	Maximum number of client connections allowed on a node. Potential issue If set too low relative to workload, connection failures may occur and node health check may fail, potentially triggering failover. Choose value considering node type and workload.

Copy parameter group

For custom parameter groups, a copy function is provided for backup purposes. Create a new parameter group by copying selected one.

1. Go to the KakaoCloud Console > Data Store > MemStore.
2. Click Parameter groups, then select Custom groups tab.
3. Select the group to copy, click [⋮] > Copy.
4. In the Copy parameter group popup, confirm input information then click Confirm.
5. On the Custom groups tab, verify the newly copied group appears.

Reset parameter group to default

Custom parameter groups offer the ability to reset values back to initial defaults. “Initial default” means the default values provided for the chosen engine version at creation. To reset means applying the initial values across all connected clusters.

info

If any connected cluster has parameter apply status Pending or Applying, resetting to default cannot be performed.

1. Go to the KakaoCloud Console > Data Store > MemStore.
2. Click Parameter groups, then select Custom groups tab.
3. Locate the group you want to reset and click [⋮] > Reset to defaults.
4. In the Reset to defaults popup, review instructions and click Confirm.
   If multiple clusters are linked, choose application method:
   - Apply in batch: Apply changes to all clusters at once.
   - Apply sequentially: Apply changes to clusters in sequence (clusters in waiting state will show Pending).
5. If clusters are connected, verify parameter apply status for each cluster.

Roll back parameter group

Custom parameter groups allow rollback to the previous parameter values after modification. This feature is only available if there is a previous modification history.

info

• If any connected cluster has parameter apply status Pending or Applying, rollback is not allowed.
• After rollback, you cannot revert to an earlier history. If you modify again, new history is created enabling a new rollback option.

1. Go to the KakaoCloud Console > Data Store > MemStore.
2. Click Parameter groups, then select Custom groups tab.
3. Find the group to rollback and click [⋮] > Rollback.
4. In the Rollback parameter group popup, review instructions and click Confirm.
5. If clusters are connected, check parameter apply status for each cluster to confirm rollback is applied.

Delete parameter group

You can delete custom parameter groups that are no longer used.

info

• Default parameter groups cannot be deleted.
• Parameter groups linked to clusters cannot be deleted.

1. Go to the KakaoCloud Console > Data Store > MemStore.
2. Click Parameter groups, then select Custom groups tab.
3. Find the group to delete and click [⋮] > Delete.
4. In the Delete parameter group popup, fill required information and click Confirm.
5. On the Custom groups tab, verify the group has been removed.

Retry parameter application

You can retry parameter application only when connected instances have apply status Error‑Sync or Suspended.
For details about apply statuses, see Parameter application status.

1. Go to the KakaoCloud Console > Data Store > MemStore.
2. Click Parameter groups, then select the Custom groups tab.
3. Select the group you want and navigate to Connected clusters tab.
4. Click the [Retry] button for the instance that needs retrying.