[tickets] [opensaf:tickets] #1031 PLM: active controller HE resets itself when first instantiating its EE
- **status**: review --> fixed - **Comment**: [staging:4db1e8] [staging:6861b7] changeset: 5877:4db1e8acdfbd branch: opensaf-4.5.x parent: 5875:b61d9d86e856 user:Alex Jones date:Fri Sep 19 21:07:24 2014 -0400 summary: plm: don't reset self during plms_ee_instantiate [#1031] changeset: 5878:6861b77dfd3b tag: tip parent: 5876:f665d08c8d5f user:Alex Jones date:Fri Sep 19 21:07:24 2014 -0400 summary: plm: don't reset self during plms_ee_instantiate [#1031] --- ** [tickets:#1031] PLM: active controller HE resets itself when first instantiating its EE** **Status:** fixed **Milestone:** 4.5.0 **Created:** Thu Aug 28, 2014 07:32 PM UTC by Alex Jones **Last Updated:** Tue Sep 02, 2014 10:23 AM UTC **Owner:** Alex Jones plms_ee_instantiate() resets the HE to which the EE belongs. But, this function is only called if there are dependencies on the HE. It is never called if there are no dependencies. If the active controller blade (modeled as an HE) has HE dependencies (some other FRUs on the board), this blade/HE will only be brought in-service when the other HEs on which it depends are brought in-service. At this point, plms_ee_instantiate() is called and the controller ends up resetting itself. This is not good! plms_ent_enable() is called in plms_move_dep_ent_to_insvc(). This function, plms_ent_enable(), is not called if there are no dependencies, and hardware is not inserted into the chassis. If the hardware is already present, and there are no dependencies, plms_ent_enable() is not called, which means plms_ee_instantiate is not called. If the active controller blade/HE does not depend on any other HEs, plms_ee_instantiate is never called, and so the blade can never reset itself. In plms_ee_instantiate() we probably need to check whether or not we are about to reset ourself, and not do it. We can safely assume that the EE is already running (since our code is executing!), and that plmcd will communicate with PLMS to set the EE states. --- Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is subscribed to https://sourceforge.net/p/opensaf/tickets/ To unsubscribe from further messages, a project admin can change settings at https://sourceforge.net/p/opensaf/admin/tickets/options. Or, if this is a mailing list, you can unsubscribe from the mailing list.-- Slashdot TV. Video for Nerds. Stuff that Matters. http://pubads.g.doubleclick.net/gampad/clk?id=160591471&iu=/4140/ostg.clktrk___ Opensaf-tickets mailing list Opensaf-tickets@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/opensaf-tickets
[tickets] [opensaf:tickets] #1031 PLM: active controller HE resets itself when first instantiating its EE
- **Comment**: I do not see the patch for review in the devellist. Please resend or paste the link to the review request. --- ** [tickets:#1031] PLM: active controller HE resets itself when first instantiating its EE** **Status:** review **Milestone:** 4.5.0 **Created:** Thu Aug 28, 2014 07:32 PM UTC by Alex Jones **Last Updated:** Thu Aug 28, 2014 09:25 PM UTC **Owner:** Alex Jones plms_ee_instantiate() resets the HE to which the EE belongs. But, this function is only called if there are dependencies on the HE. It is never called if there are no dependencies. If the active controller blade (modeled as an HE) has HE dependencies (some other FRUs on the board), this blade/HE will only be brought in-service when the other HEs on which it depends are brought in-service. At this point, plms_ee_instantiate() is called and the controller ends up resetting itself. This is not good! plms_ent_enable() is called in plms_move_dep_ent_to_insvc(). This function, plms_ent_enable(), is not called if there are no dependencies, and hardware is not inserted into the chassis. If the hardware is already present, and there are no dependencies, plms_ent_enable() is not called, which means plms_ee_instantiate is not called. If the active controller blade/HE does not depend on any other HEs, plms_ee_instantiate is never called, and so the blade can never reset itself. In plms_ee_instantiate() we probably need to check whether or not we are about to reset ourself, and not do it. We can safely assume that the EE is already running (since our code is executing!), and that plmcd will communicate with PLMS to set the EE states. --- Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is subscribed to https://sourceforge.net/p/opensaf/tickets/ To unsubscribe from further messages, a project admin can change settings at https://sourceforge.net/p/opensaf/admin/tickets/options. Or, if this is a mailing list, you can unsubscribe from the mailing list.-- Slashdot TV. Video for Nerds. Stuff that matters. http://tv.slashdot.org/___ Opensaf-tickets mailing list Opensaf-tickets@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/opensaf-tickets
[tickets] [opensaf:tickets] #1031 PLM: active controller HE resets itself when first instantiating its EE
- **status**: accepted --> review --- ** [tickets:#1031] PLM: active controller HE resets itself when first instantiating its EE** **Status:** review **Milestone:** 4.5.0 **Created:** Thu Aug 28, 2014 07:32 PM UTC by Alex Jones **Last Updated:** Thu Aug 28, 2014 07:32 PM UTC **Owner:** Alex Jones plms_ee_instantiate() resets the HE to which the EE belongs. But, this function is only called if there are dependencies on the HE. It is never called if there are no dependencies. If the active controller blade (modeled as an HE) has HE dependencies (some other FRUs on the board), this blade/HE will only be brought in-service when the other HEs on which it depends are brought in-service. At this point, plms_ee_instantiate() is called and the controller ends up resetting itself. This is not good! plms_ent_enable() is called in plms_move_dep_ent_to_insvc(). This function, plms_ent_enable(), is not called if there are no dependencies, and hardware is not inserted into the chassis. If the hardware is already present, and there are no dependencies, plms_ent_enable() is not called, which means plms_ee_instantiate is not called. If the active controller blade/HE does not depend on any other HEs, plms_ee_instantiate is never called, and so the blade can never reset itself. In plms_ee_instantiate() we probably need to check whether or not we are about to reset ourself, and not do it. We can safely assume that the EE is already running (since our code is executing!), and that plmcd will communicate with PLMS to set the EE states. --- Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is subscribed to https://sourceforge.net/p/opensaf/tickets/ To unsubscribe from further messages, a project admin can change settings at https://sourceforge.net/p/opensaf/admin/tickets/options. Or, if this is a mailing list, you can unsubscribe from the mailing list.-- Slashdot TV. Video for Nerds. Stuff that matters. http://tv.slashdot.org/___ Opensaf-tickets mailing list Opensaf-tickets@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/opensaf-tickets
[tickets] [opensaf:tickets] #1031 PLM: active controller HE resets itself when first instantiating its EE
--- ** [tickets:#1031] PLM: active controller HE resets itself when first instantiating its EE** **Status:** accepted **Milestone:** 4.5.0 **Created:** Thu Aug 28, 2014 07:32 PM UTC by Alex Jones **Last Updated:** Thu Aug 28, 2014 07:32 PM UTC **Owner:** Alex Jones plms_ee_instantiate() resets the HE to which the EE belongs. But, this function is only called if there are dependencies on the HE. It is never called if there are no dependencies. If the active controller blade (modeled as an HE) has HE dependencies (some other FRUs on the board), this blade/HE will only be brought in-service when the other HEs on which it depends are brought in-service. At this point, plms_ee_instantiate() is called and the controller ends up resetting itself. This is not good! plms_ent_enable() is called in plms_move_dep_ent_to_insvc(). This function, plms_ent_enable(), is not called if there are no dependencies, and hardware is not inserted into the chassis. If the hardware is already present, and there are no dependencies, plms_ent_enable() is not called, which means plms_ee_instantiate is not called. If the active controller blade/HE does not depend on any other HEs, plms_ee_instantiate is never called, and so the blade can never reset itself. In plms_ee_instantiate() we probably need to check whether or not we are about to reset ourself, and not do it. We can safely assume that the EE is already running (since our code is executing!), and that plmcd will communicate with PLMS to set the EE states. --- Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is subscribed to https://sourceforge.net/p/opensaf/tickets/ To unsubscribe from further messages, a project admin can change settings at https://sourceforge.net/p/opensaf/admin/tickets/options. Or, if this is a mailing list, you can unsubscribe from the mailing list.-- Slashdot TV. Video for Nerds. Stuff that matters. http://tv.slashdot.org/___ Opensaf-tickets mailing list Opensaf-tickets@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/opensaf-tickets
