[jira] Commented: (IBATIS-508) Lock occurs when heavy traffic
[
https://issues.apache.org/jira/browse/IBATIS-508?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12685362#action_12685362
]
Clinton Begin commented on IBATIS-508:
--
Kai, you have made it through... I've just been swamped with stuff in the last
few months, and unfortunately dropped the ball. I'll get on this over the
weekend.
Cheers,
Clinton
> Lock occurs when heavy traffic
> --
>
> Key: IBATIS-508
> URL: https://issues.apache.org/jira/browse/IBATIS-508
> Project: iBatis for Java
> Issue Type: Bug
> Components: SQL Maps
>Affects Versions: 2.3.0
> Environment: all platform
>Reporter: Takayuki Ikeda
>Priority: Minor
> Attachments: iBATIS-508.patch
>
>
> In heavy traffic , lock occurs at
> com.ibatis.common.beans.ClassInfo#getInstance in case of cacheEnabled = true.
> It seems to be happend because of "double synchronized"
> ClassInfo.java
> ---
> private static final Map CLASS_INFO_MAP = Collections.synchronizedMap(new
> HashMap());
> public static ClassInfo getInstance(Class clazz) {
> if (cacheEnabled) {
> synchronized (clazz) {
> ---
> I think we get more concurrency to access getInstance if the synchronized
> clause in getInstance method is deleted.
> There is no side-effect to do this.
> #just created several useless ClassInfo objects :-)
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (IBATIS-508) Lock occurs when heavy traffic
[
https://issues.apache.org/jira/browse/IBATIS-508?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12684015#action_12684015
]
Kai Grabfelder commented on IBATIS-508:
---
wow it sounds like it would be very reasonable to apply this patch. @Clinton or
any over dev. could you apply it? Unfortunately it seems that I did not made it
through the pmc commiter approval process ;-(
> Lock occurs when heavy traffic
> --
>
> Key: IBATIS-508
> URL: https://issues.apache.org/jira/browse/IBATIS-508
> Project: iBatis for Java
> Issue Type: Bug
> Components: SQL Maps
>Affects Versions: 2.3.0
> Environment: all platform
>Reporter: Takayuki Ikeda
>Priority: Minor
> Attachments: iBATIS-508.patch
>
>
> In heavy traffic , lock occurs at
> com.ibatis.common.beans.ClassInfo#getInstance in case of cacheEnabled = true.
> It seems to be happend because of "double synchronized"
> ClassInfo.java
> ---
> private static final Map CLASS_INFO_MAP = Collections.synchronizedMap(new
> HashMap());
> public static ClassInfo getInstance(Class clazz) {
> if (cacheEnabled) {
> synchronized (clazz) {
> ---
> I think we get more concurrency to access getInstance if the synchronized
> clause in getInstance method is deleted.
> There is no side-effect to do this.
> #just created several useless ClassInfo objects :-)
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (IBATIS-508) Lock occurs when heavy traffic
[
https://issues.apache.org/jira/browse/IBATIS-508?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12683961#action_12683961
]
Brian Lomeland commented on IBATIS-508:
---
Forgot to add that IBATIS-508 affects iBATIS 2.3.4 and this patch is for the
latest SVN version of the ClassInfo.java.
> Lock occurs when heavy traffic
> --
>
> Key: IBATIS-508
> URL: https://issues.apache.org/jira/browse/IBATIS-508
> Project: iBatis for Java
> Issue Type: Bug
> Components: SQL Maps
>Affects Versions: 2.3.0
> Environment: all platform
>Reporter: Takayuki Ikeda
>Priority: Minor
> Attachments: iBATIS-508.patch
>
>
> In heavy traffic , lock occurs at
> com.ibatis.common.beans.ClassInfo#getInstance in case of cacheEnabled = true.
> It seems to be happend because of "double synchronized"
> ClassInfo.java
> ---
> private static final Map CLASS_INFO_MAP = Collections.synchronizedMap(new
> HashMap());
> public static ClassInfo getInstance(Class clazz) {
> if (cacheEnabled) {
> synchronized (clazz) {
> ---
> I think we get more concurrency to access getInstance if the synchronized
> clause in getInstance method is deleted.
> There is no side-effect to do this.
> #just created several useless ClassInfo objects :-)
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
