Re: [openstack-dev] [Fuel] Assigning VIPs on network config serialization
Hi all, Let's summarize what we have by now. 1/ We agreed on removing *VIP allocation* code from GET network_configuration API handler. Already allocated VIPs should be returned by this call though. 2/ We agreed on checking whether it's enough IP addressed in the pool as a part of Verify Network API call (currently we do it only in before deployment checks). 3/ We should keep in mind (and perhaps make some ticket) about ability to allocate VIPs on demand.. on some API call, or even be able to assign custom one. I think we need update LP ticket of (1), and file a ticket for (2). Sounds like a plan, ha? Thanks, Igor On Wed, Oct 21, 2015 at 3:54 PM, Aleksey Kasatkin wrote: >> Then we should close [1] as invalid, shoudn’t we? > > AFAIC, no. We should check whether there is enough IPs for nodes / > VIPs with current network configuration. > > I'd propose to add a handler for allocation of VIPs if VIPs can be useful > before deployment. > I'm not sure what are the cases. > > > > Aleksey Kasatkin > > > On Wed, Oct 21, 2015 at 11:45 AM, Roman Prykhodchenko wrote: >> >> Then we should close [1] as invalid, shoudn’t we? >> >> > 20 жовт. 2015 р. о 15:55 Igor Kalnitsky >> > написав(ла): >> > >> > Roman, >> > >> >> This behavior is actually described in [1]. Should we allocate >> >> VIPs on network check as well? >> > >> > No, we shouldn't. We should check whether it's enough IPs for nodes / >> > VIPs with current network configuration, but no more. >> > >> > - igor >> > >> > On Tue, Oct 20, 2015 at 4:54 PM, Igor Kalnitsky >> > wrote: >> >> Andrew, >> >> >> >>> but the problem lies that VIP's need to already be allocated. >> >> >> >> Why? Fuel doesn't use them on this stage. >> >> >> >> >> >>> They need to be allocated on network update, or when a node role >> >>> requiring >> >>> one is added to the environment. >> >> >> >> It looks like either you or me misunderstood something. AFAIK, node >> >> role itself has nothing in common with VIPs. It doesn't require any of >> >> them. >> >> >> >> Currently VIPs are requested by network roles, and network roles are >> >> the same for all nodes (except Network Templating case). Network roles >> >> are assigned on network, and if VIP is required for network role it >> >> will be allocated in the assigned network. >> >> >> >> So basically, it requires a huge effort to redesign our allocation >> >> system to achieve what you want, because: >> >> >> >> * Each time you reassign network role, the corresponding VIP should be >> >> re-allocated in the database either. >> >> * Each time you enable/disable plugins, the VIPs should be >> >> re-allocated, because plugins may export network roles. >> >> * Each time you add new node to cluster, the VIPs should be >> >> re-allocated, because with new node you simply may run out of free >> >> IPs. And, btw, should we assign IP on added nodes here? Or maybe >> >> postpone to serialization step? >> >> >> >> Well, Andrew, I believe we don't have enough resources to implement >> >> your proposal. Moreover, the proposed approach requires a lot of >> >> discussions and design meetings. And it definitely should be >> >> implemented in scope of some blueprint, not a bugfix. >> >> >> >> >> >>> Not knowing the address until serialization for deployment is too >> >>> late. >> >> >> >> Once again - why? I agree, perhaps it would be useful, but there's no >> >> strict requirement on this and we should resolve our issues >> >> step-by-step. See my response above. >> >> >> >> >> >>> No, Again, this is too late. >> >> >> >> Too late for what? >> >> >> >> >> >> - Igor >> >> >> >> On Tue, Oct 20, 2015 at 12:38 PM, Roman Prykhodchenko >> >> wrote: >> >>> My concern here is that there is also a Network check feature that >> >>> according to its name should check things like whether or not there’s >> >>> enough >> >>> IP addresses in all ranges in a network. The problem is that it may be >> >>> run >> >>> at any time, even when VIPs are not yet allocated. From a user-side the >> >>> workflow may look a little wrong: >> >>> >> >>> 1. Check network => get "Everything is fine" >> >>> 2. Right after that press Apply Changes => get "Network configuration >> >>> is bad" >> >>> >> >>> This behavior is actually described in [1]. Should we allocate VIPs on >> >>> network check as well? >> >>> >> >>> >> >>> 1. https://bugs.launchpad.net/fuel/+bug/1487996 >> >>> >> >>> >> >>> - romcheg >> >>> >> >>> >> 19 жовт. 2015 р. о 18:28 Igor Kalnitsky >> написав(ла): >> >> Hi Roman, >> >> > Not assign addresses to VIPs is a network configuration is being >> > serialized for API output. >> >> AFAIK, that's not truth. Fuel UI and OSTF relies only on *public* >> VIP. >> So we can keep only *public* VIP, and do not assign / show others. >> >> > Check number of IP addresses wherever it is possible to "spoil" >> > network >> > configuration: when a role get’s assigned to a node, when network >> > change
Re: [openstack-dev] [Fuel] Assigning VIPs on network config serialization
> Then we should close [1] as invalid, shoudn’t we? AFAIC, no. We should check whether there is enough IPs for nodes / VIPs with current network configuration. I'd propose to add a handler for allocation of VIPs if VIPs can be useful before deployment. I'm not sure what are the cases. Aleksey Kasatkin On Wed, Oct 21, 2015 at 11:45 AM, Roman Prykhodchenko wrote: > Then we should close [1] as invalid, shoudn’t we? > > > 20 жовт. 2015 р. о 15:55 Igor Kalnitsky > написав(ла): > > > > Roman, > > > >> This behavior is actually described in [1]. Should we allocate > >> VIPs on network check as well? > > > > No, we shouldn't. We should check whether it's enough IPs for nodes / > > VIPs with current network configuration, but no more. > > > > - igor > > > > On Tue, Oct 20, 2015 at 4:54 PM, Igor Kalnitsky > wrote: > >> Andrew, > >> > >>> but the problem lies that VIP's need to already be allocated. > >> > >> Why? Fuel doesn't use them on this stage. > >> > >> > >>> They need to be allocated on network update, or when a node role > requiring > >>> one is added to the environment. > >> > >> It looks like either you or me misunderstood something. AFAIK, node > >> role itself has nothing in common with VIPs. It doesn't require any of > >> them. > >> > >> Currently VIPs are requested by network roles, and network roles are > >> the same for all nodes (except Network Templating case). Network roles > >> are assigned on network, and if VIP is required for network role it > >> will be allocated in the assigned network. > >> > >> So basically, it requires a huge effort to redesign our allocation > >> system to achieve what you want, because: > >> > >> * Each time you reassign network role, the corresponding VIP should be > >> re-allocated in the database either. > >> * Each time you enable/disable plugins, the VIPs should be > >> re-allocated, because plugins may export network roles. > >> * Each time you add new node to cluster, the VIPs should be > >> re-allocated, because with new node you simply may run out of free > >> IPs. And, btw, should we assign IP on added nodes here? Or maybe > >> postpone to serialization step? > >> > >> Well, Andrew, I believe we don't have enough resources to implement > >> your proposal. Moreover, the proposed approach requires a lot of > >> discussions and design meetings. And it definitely should be > >> implemented in scope of some blueprint, not a bugfix. > >> > >> > >>> Not knowing the address until serialization for deployment is too late. > >> > >> Once again - why? I agree, perhaps it would be useful, but there's no > >> strict requirement on this and we should resolve our issues > >> step-by-step. See my response above. > >> > >> > >>> No, Again, this is too late. > >> > >> Too late for what? > >> > >> > >> - Igor > >> > >> On Tue, Oct 20, 2015 at 12:38 PM, Roman Prykhodchenko > wrote: > >>> My concern here is that there is also a Network check feature that > according to its name should check things like whether or not there’s > enough IP addresses in all ranges in a network. The problem is that it may > be run at any time, even when VIPs are not yet allocated. From a user-side > the workflow may look a little wrong: > >>> > >>> 1. Check network => get "Everything is fine" > >>> 2. Right after that press Apply Changes => get "Network configuration > is bad" > >>> > >>> This behavior is actually described in [1]. Should we allocate VIPs on > network check as well? > >>> > >>> > >>> 1. https://bugs.launchpad.net/fuel/+bug/1487996 > >>> > >>> > >>> - romcheg > >>> > >>> > 19 жовт. 2015 р. о 18:28 Igor Kalnitsky > написав(ла): > > Hi Roman, > > > Not assign addresses to VIPs is a network configuration is being > > serialized for API output. > > AFAIK, that's not truth. Fuel UI and OSTF relies only on *public* VIP. > So we can keep only *public* VIP, and do not assign / show others. > > > Check number of IP addresses wherever it is possible to "spoil" > network > > configuration: when a role get’s assigned to a node, when network > > changes or network templates are applied. > > It won't work that way. What if you enable plugin when all roles are > assigned? What if you change networks, and now it's not enough IPs? Or > what if enable plugin that requires 5 VIPs in public network by > default, and it's not enough. But by using network templates you > assign this netrole to management network? > > From what I can say the proposed approach requires to put checks > here-and-there around the code. Let's do not overcomplicate things > without real need. > > So let me share my thoughts regarding this issue. > > * We shouldn't *allocate* VIPs when we make GET request on network > configuration handler. It should return only *already allocated* VIPs > and no more. > * VIP allocation should be performed when we run deployment. > * Befor
Re: [openstack-dev] [Fuel] Assigning VIPs on network config serialization
Then we should close [1] as invalid, shoudn’t we? > 20 жовт. 2015 р. о 15:55 Igor Kalnitsky написав(ла): > > Roman, > >> This behavior is actually described in [1]. Should we allocate >> VIPs on network check as well? > > No, we shouldn't. We should check whether it's enough IPs for nodes / > VIPs with current network configuration, but no more. > > - igor > > On Tue, Oct 20, 2015 at 4:54 PM, Igor Kalnitsky > wrote: >> Andrew, >> >>> but the problem lies that VIP's need to already be allocated. >> >> Why? Fuel doesn't use them on this stage. >> >> >>> They need to be allocated on network update, or when a node role requiring >>> one is added to the environment. >> >> It looks like either you or me misunderstood something. AFAIK, node >> role itself has nothing in common with VIPs. It doesn't require any of >> them. >> >> Currently VIPs are requested by network roles, and network roles are >> the same for all nodes (except Network Templating case). Network roles >> are assigned on network, and if VIP is required for network role it >> will be allocated in the assigned network. >> >> So basically, it requires a huge effort to redesign our allocation >> system to achieve what you want, because: >> >> * Each time you reassign network role, the corresponding VIP should be >> re-allocated in the database either. >> * Each time you enable/disable plugins, the VIPs should be >> re-allocated, because plugins may export network roles. >> * Each time you add new node to cluster, the VIPs should be >> re-allocated, because with new node you simply may run out of free >> IPs. And, btw, should we assign IP on added nodes here? Or maybe >> postpone to serialization step? >> >> Well, Andrew, I believe we don't have enough resources to implement >> your proposal. Moreover, the proposed approach requires a lot of >> discussions and design meetings. And it definitely should be >> implemented in scope of some blueprint, not a bugfix. >> >> >>> Not knowing the address until serialization for deployment is too late. >> >> Once again - why? I agree, perhaps it would be useful, but there's no >> strict requirement on this and we should resolve our issues >> step-by-step. See my response above. >> >> >>> No, Again, this is too late. >> >> Too late for what? >> >> >> - Igor >> >> On Tue, Oct 20, 2015 at 12:38 PM, Roman Prykhodchenko >> wrote: >>> My concern here is that there is also a Network check feature that >>> according to its name should check things like whether or not there’s >>> enough IP addresses in all ranges in a network. The problem is that it may >>> be run at any time, even when VIPs are not yet allocated. From a user-side >>> the workflow may look a little wrong: >>> >>> 1. Check network => get "Everything is fine" >>> 2. Right after that press Apply Changes => get "Network configuration is >>> bad" >>> >>> This behavior is actually described in [1]. Should we allocate VIPs on >>> network check as well? >>> >>> >>> 1. https://bugs.launchpad.net/fuel/+bug/1487996 >>> >>> >>> - romcheg >>> >>> 19 жовт. 2015 р. о 18:28 Igor Kalnitsky написав(ла): Hi Roman, > Not assign addresses to VIPs is a network configuration is being > serialized for API output. AFAIK, that's not truth. Fuel UI and OSTF relies only on *public* VIP. So we can keep only *public* VIP, and do not assign / show others. > Check number of IP addresses wherever it is possible to "spoil" network > configuration: when a role get’s assigned to a node, when network > changes or network templates are applied. It won't work that way. What if you enable plugin when all roles are assigned? What if you change networks, and now it's not enough IPs? Or what if enable plugin that requires 5 VIPs in public network by default, and it's not enough. But by using network templates you assign this netrole to management network? From what I can say the proposed approach requires to put checks here-and-there around the code. Let's do not overcomplicate things without real need. So let me share my thoughts regarding this issue. * We shouldn't *allocate* VIPs when we make GET request on network configuration handler. It should return only *already allocated* VIPs and no more. * VIP allocation should be performed when we run deployment. * Before deployment checks should fail, if there's not enough VIPs or other resources. So users fix them, and try again. * Both Fuel UI and OSTF needs VIPs only when cluster is deployed, and it's safe to return allocated VIPs on that stage. So what do you think guys? Thanks, Igor On Fri, Oct 16, 2015 at 5:25 PM, Roman Prykhodchenko wrote: > Hi folks! > > I’ve been discussing several bugs [1-3] with some folks and noticed that > they share the same root cause whic
Re: [openstack-dev] [Fuel] Assigning VIPs on network config serialization
Roman, > This behavior is actually described in [1]. Should we allocate > VIPs on network check as well? No, we shouldn't. We should check whether it's enough IPs for nodes / VIPs with current network configuration, but no more. - igor On Tue, Oct 20, 2015 at 4:54 PM, Igor Kalnitsky wrote: > Andrew, > >> but the problem lies that VIP's need to already be allocated. > > Why? Fuel doesn't use them on this stage. > > >> They need to be allocated on network update, or when a node role requiring >> one is added to the environment. > > It looks like either you or me misunderstood something. AFAIK, node > role itself has nothing in common with VIPs. It doesn't require any of > them. > > Currently VIPs are requested by network roles, and network roles are > the same for all nodes (except Network Templating case). Network roles > are assigned on network, and if VIP is required for network role it > will be allocated in the assigned network. > > So basically, it requires a huge effort to redesign our allocation > system to achieve what you want, because: > > * Each time you reassign network role, the corresponding VIP should be > re-allocated in the database either. > * Each time you enable/disable plugins, the VIPs should be > re-allocated, because plugins may export network roles. > * Each time you add new node to cluster, the VIPs should be > re-allocated, because with new node you simply may run out of free > IPs. And, btw, should we assign IP on added nodes here? Or maybe > postpone to serialization step? > > Well, Andrew, I believe we don't have enough resources to implement > your proposal. Moreover, the proposed approach requires a lot of > discussions and design meetings. And it definitely should be > implemented in scope of some blueprint, not a bugfix. > > >> Not knowing the address until serialization for deployment is too late. > > Once again - why? I agree, perhaps it would be useful, but there's no > strict requirement on this and we should resolve our issues > step-by-step. See my response above. > > >> No, Again, this is too late. > > Too late for what? > > > - Igor > > On Tue, Oct 20, 2015 at 12:38 PM, Roman Prykhodchenko wrote: >> My concern here is that there is also a Network check feature that according >> to its name should check things like whether or not there’s enough IP >> addresses in all ranges in a network. The problem is that it may be run at >> any time, even when VIPs are not yet allocated. From a user-side the >> workflow may look a little wrong: >> >> 1. Check network => get "Everything is fine" >> 2. Right after that press Apply Changes => get "Network configuration is >> bad" >> >> This behavior is actually described in [1]. Should we allocate VIPs on >> network check as well? >> >> >> 1. https://bugs.launchpad.net/fuel/+bug/1487996 >> >> >> - romcheg >> >> >>> 19 жовт. 2015 р. о 18:28 Igor Kalnitsky >>> написав(ла): >>> >>> Hi Roman, >>> Not assign addresses to VIPs is a network configuration is being serialized for API output. >>> >>> AFAIK, that's not truth. Fuel UI and OSTF relies only on *public* VIP. >>> So we can keep only *public* VIP, and do not assign / show others. >>> Check number of IP addresses wherever it is possible to "spoil" network configuration: when a role get’s assigned to a node, when network changes or network templates are applied. >>> >>> It won't work that way. What if you enable plugin when all roles are >>> assigned? What if you change networks, and now it's not enough IPs? Or >>> what if enable plugin that requires 5 VIPs in public network by >>> default, and it's not enough. But by using network templates you >>> assign this netrole to management network? >>> >>> From what I can say the proposed approach requires to put checks >>> here-and-there around the code. Let's do not overcomplicate things >>> without real need. >>> >>> So let me share my thoughts regarding this issue. >>> >>> * We shouldn't *allocate* VIPs when we make GET request on network >>> configuration handler. It should return only *already allocated* VIPs >>> and no more. >>> * VIP allocation should be performed when we run deployment. >>> * Before deployment checks should fail, if there's not enough VIPs or >>> other resources. So users fix them, and try again. >>> * Both Fuel UI and OSTF needs VIPs only when cluster is deployed, and >>> it's safe to return allocated VIPs on that stage. >>> >>> So what do you think guys? >>> >>> Thanks, >>> Igor >>> >>> On Fri, Oct 16, 2015 at 5:25 PM, Roman Prykhodchenko >>> wrote: Hi folks! I’ve been discussing several bugs [1-3] with some folks and noticed that they share the same root cause which is that network serialization fails, if there’s not enough IP addresses in all available ranges of one of the available networks to assign them to all VIPs. There are several possible solutions for this issue: a. Not assign addresses to VIPs is a network configu
Re: [openstack-dev] [Fuel] Assigning VIPs on network config serialization
Andrew, > but the problem lies that VIP's need to already be allocated. Why? Fuel doesn't use them on this stage. > They need to be allocated on network update, or when a node role requiring > one is added to the environment. It looks like either you or me misunderstood something. AFAIK, node role itself has nothing in common with VIPs. It doesn't require any of them. Currently VIPs are requested by network roles, and network roles are the same for all nodes (except Network Templating case). Network roles are assigned on network, and if VIP is required for network role it will be allocated in the assigned network. So basically, it requires a huge effort to redesign our allocation system to achieve what you want, because: * Each time you reassign network role, the corresponding VIP should be re-allocated in the database either. * Each time you enable/disable plugins, the VIPs should be re-allocated, because plugins may export network roles. * Each time you add new node to cluster, the VIPs should be re-allocated, because with new node you simply may run out of free IPs. And, btw, should we assign IP on added nodes here? Or maybe postpone to serialization step? Well, Andrew, I believe we don't have enough resources to implement your proposal. Moreover, the proposed approach requires a lot of discussions and design meetings. And it definitely should be implemented in scope of some blueprint, not a bugfix. > Not knowing the address until serialization for deployment is too late. Once again - why? I agree, perhaps it would be useful, but there's no strict requirement on this and we should resolve our issues step-by-step. See my response above. > No, Again, this is too late. Too late for what? - Igor On Tue, Oct 20, 2015 at 12:38 PM, Roman Prykhodchenko wrote: > My concern here is that there is also a Network check feature that according > to its name should check things like whether or not there’s enough IP > addresses in all ranges in a network. The problem is that it may be run at > any time, even when VIPs are not yet allocated. From a user-side the workflow > may look a little wrong: > > 1. Check network => get "Everything is fine" > 2. Right after that press Apply Changes => get "Network configuration is bad" > > This behavior is actually described in [1]. Should we allocate VIPs on > network check as well? > > > 1. https://bugs.launchpad.net/fuel/+bug/1487996 > > > - romcheg > > >> 19 жовт. 2015 р. о 18:28 Igor Kalnitsky >> написав(ла): >> >> Hi Roman, >> >>> Not assign addresses to VIPs is a network configuration is being >>> serialized for API output. >> >> AFAIK, that's not truth. Fuel UI and OSTF relies only on *public* VIP. >> So we can keep only *public* VIP, and do not assign / show others. >> >>> Check number of IP addresses wherever it is possible to "spoil" network >>> configuration: when a role get’s assigned to a node, when network >>> changes or network templates are applied. >> >> It won't work that way. What if you enable plugin when all roles are >> assigned? What if you change networks, and now it's not enough IPs? Or >> what if enable plugin that requires 5 VIPs in public network by >> default, and it's not enough. But by using network templates you >> assign this netrole to management network? >> >> From what I can say the proposed approach requires to put checks >> here-and-there around the code. Let's do not overcomplicate things >> without real need. >> >> So let me share my thoughts regarding this issue. >> >> * We shouldn't *allocate* VIPs when we make GET request on network >> configuration handler. It should return only *already allocated* VIPs >> and no more. >> * VIP allocation should be performed when we run deployment. >> * Before deployment checks should fail, if there's not enough VIPs or >> other resources. So users fix them, and try again. >> * Both Fuel UI and OSTF needs VIPs only when cluster is deployed, and >> it's safe to return allocated VIPs on that stage. >> >> So what do you think guys? >> >> Thanks, >> Igor >> >> On Fri, Oct 16, 2015 at 5:25 PM, Roman Prykhodchenko wrote: >>> Hi folks! >>> >>> I’ve been discussing several bugs [1-3] with some folks and noticed that >>> they share the same root cause which is that network serialization fails, >>> if there’s not enough IP addresses in all available ranges of one of the >>> available networks to assign them to all VIPs. There are several possible >>> solutions for this issue: >>> >>> a. Not assign addresses to VIPs is a network configuration is being >>> serialized for API output. >>> A lot of external tools and modules, i. e., OSTF, rely on that information >>> so this relatively small change in Nailgun will require big >>> cross-components changes. Therefore this change can only be done as a >>> feature but it seems to be the way this issue must be fixed. >>> >>> b. Leave some VIPs without IP addresses >>> If network configuration is generated for API output it is possible to
Re: [openstack-dev] [Fuel] Assigning VIPs on network config serialization
My concern here is that there is also a Network check feature that according to its name should check things like whether or not there’s enough IP addresses in all ranges in a network. The problem is that it may be run at any time, even when VIPs are not yet allocated. From a user-side the workflow may look a little wrong: 1. Check network => get "Everything is fine" 2. Right after that press Apply Changes => get "Network configuration is bad" This behavior is actually described in [1]. Should we allocate VIPs on network check as well? 1. https://bugs.launchpad.net/fuel/+bug/1487996 - romcheg > 19 жовт. 2015 р. о 18:28 Igor Kalnitsky написав(ла): > > Hi Roman, > >> Not assign addresses to VIPs is a network configuration is being >> serialized for API output. > > AFAIK, that's not truth. Fuel UI and OSTF relies only on *public* VIP. > So we can keep only *public* VIP, and do not assign / show others. > >> Check number of IP addresses wherever it is possible to "spoil" network >> configuration: when a role get’s assigned to a node, when network >> changes or network templates are applied. > > It won't work that way. What if you enable plugin when all roles are > assigned? What if you change networks, and now it's not enough IPs? Or > what if enable plugin that requires 5 VIPs in public network by > default, and it's not enough. But by using network templates you > assign this netrole to management network? > > From what I can say the proposed approach requires to put checks > here-and-there around the code. Let's do not overcomplicate things > without real need. > > So let me share my thoughts regarding this issue. > > * We shouldn't *allocate* VIPs when we make GET request on network > configuration handler. It should return only *already allocated* VIPs > and no more. > * VIP allocation should be performed when we run deployment. > * Before deployment checks should fail, if there's not enough VIPs or > other resources. So users fix them, and try again. > * Both Fuel UI and OSTF needs VIPs only when cluster is deployed, and > it's safe to return allocated VIPs on that stage. > > So what do you think guys? > > Thanks, > Igor > > On Fri, Oct 16, 2015 at 5:25 PM, Roman Prykhodchenko wrote: >> Hi folks! >> >> I’ve been discussing several bugs [1-3] with some folks and noticed that >> they share the same root cause which is that network serialization fails, if >> there’s not enough IP addresses in all available ranges of one of the >> available networks to assign them to all VIPs. There are several possible >> solutions for this issue: >> >> a. Not assign addresses to VIPs is a network configuration is being >> serialized for API output. >> A lot of external tools and modules, i. e., OSTF, rely on that information >> so this relatively small change in Nailgun will require big cross-components >> changes. Therefore this change can only be done as a feature but it seems to >> be the way this issue must be fixed. >> >> b. Leave some VIPs without IP addresses >> If network configuration is generated for API output it is possible to leave >> some VIPs without IP addresses assigned. This will only create more mess >> around Nailgun and bring more issues that it will resolve. >> >> c. Check number of IP addresses wherever it is possible to "spoil" network >> configuration: when a role get’s assigned to a node, when network changes or >> network templates are applied. >> >> >> The proposal is to follow [c] as a fast solution and file a blueprint for >> [a]. Opinions? >> >> >> 1 https://bugs.launchpad.net/fuel/+bug/1487996 >> 2 https://bugs.launchpad.net/fuel/+bug/1500394 >> 3 https://bugs.launchpad.net/fuel/+bug/1504572 >> >> >> - romcheg >> >> __ >> OpenStack Development Mailing List (not for usage questions) >> Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe >> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev >> > > __ > OpenStack Development Mailing List (not for usage questions) > Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev signature.asc Description: Message signed with OpenPGP using GPGMail __ OpenStack Development Mailing List (not for usage questions) Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [Fuel] Assigning VIPs on network config serialization
On Mon, Oct 19, 2015 at 9:29 AM Igor Kalnitsky wrote: > Hi Roman, > > > Not assign addresses to VIPs is a network configuration is being > > serialized for API output. > > AFAIK, that's not truth. Fuel UI and OSTF relies only on *public* VIP. > So we can keep only *public* VIP, and do not assign / show others. > > > Check number of IP addresses wherever it is possible to "spoil" network > > configuration: when a role get’s assigned to a node, when network > > changes or network templates are applied. > > It won't work that way. What if you enable plugin when all roles are > assigned? What if you change networks, and now it's not enough IPs? Or > what if enable plugin that requires 5 VIPs in public network by > default, and it's not enough. But by using network templates you > assign this netrole to management network? > > From what I can say the proposed approach requires to put checks > here-and-there around the code. Let's do not overcomplicate things > without real need. > > So let me share my thoughts regarding this issue. > > * We shouldn't *allocate* VIPs when we make GET request on network > configuration handler. It should return only *already allocated* VIPs > and no more. I agree, GET handler should not allocate vips, but the problem lies that VIP's need to already be allocated. They need to be allocated on network update, or when a node role requiring one is added to the environment. Not knowing the address until serialization for deployment is too late. The user needs feedback on the network page. 1) They need to know how many vip's are on each network (eventually they should be able to set these, but that is slightly off topic) 2) When saving a change they need to know that they have ENOUGH IP's for ALL of the currently configured VIPs and Nodes. Clicking deploy and getting "You don't have enough addresses start over" with out any feedback on why is not valuable to the user and they go back to settings pages, count their nodes, and then are "yup, there are enough, fuel is broken". Further compacting the issue, now plugins and tasks are going to start to be able to configure more VIP addresses. In this case it will become even more confusing as to how many VIP addresses (and from which network) they come from with out good feedback up front. The only way we can do this is generating the VIP addresses as soon as we have enough data for them, not at deployment serialization. * VIP allocation should be performed when we run deployment. > No, see above > * Before deployment checks should fail, if there's not enough VIPs or > other resources. So users fix them, and try again. > No, see above, its a bad User eXperience (UX) and needs to be stopped. This is why Roman raised the issue. > * Both Fuel UI and OSTF needs VIPs only when cluster is deployed, and > it's safe to return allocated VIPs on that stage. > No, Again, this is too late. > > So what do you think guys? > > Thanks, > Igor > > On Fri, Oct 16, 2015 at 5:25 PM, Roman Prykhodchenko > wrote: > > Hi folks! > > > > I’ve been discussing several bugs [1-3] with some folks and noticed that > they share the same root cause which is that network serialization fails, > if there’s not enough IP addresses in all available ranges of one of the > available networks to assign them to all VIPs. There are several possible > solutions for this issue: > > > > a. Not assign addresses to VIPs is a network configuration is being > serialized for API output. > > A lot of external tools and modules, i. e., OSTF, rely on that > information so this relatively small change in Nailgun will require big > cross-components changes. Therefore this change can only be done as a > feature but it seems to be the way this issue must be fixed. > > > > b. Leave some VIPs without IP addresses > > If network configuration is generated for API output it is possible to > leave some VIPs without IP addresses assigned. This will only create more > mess around Nailgun and bring more issues that it will resolve. > > > > c. Check number of IP addresses wherever it is possible to "spoil" > network configuration: when a role get’s assigned to a node, when network > changes or network templates are applied. > > > > > > The proposal is to follow [c] as a fast solution and file a blueprint > for [a]. Opinions? > > > > > > 1 https://bugs.launchpad.net/fuel/+bug/1487996 > > 2 https://bugs.launchpad.net/fuel/+bug/1500394 > > 3 https://bugs.launchpad.net/fuel/+bug/1504572 > > > > > > - romcheg > > > > > __ > > OpenStack Development Mailing List (not for usage questions) > > Unsubscribe: > openstack-dev-requ...@lists.openstack.org?subject:unsubscribe > > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev > > > > __ > OpenStack Development Mailing List (not for usage questions) > Unsubscribe: openstack-dev-requ...@lists.openstack.
Re: [openstack-dev] [Fuel] Assigning VIPs on network config serialization
Aleksey, Thank you for input. > We have a ticket for 8.0 to give an ability to set arbitrary addresses for > VIPs That shouldn't affect proposed approach. The main idea is to return *allocated VIPs* but *do not allocate* them implicitly. > we also need to provide an ability to run CheckBeforeDeployment task > separately That's another story. Let's keep close to main subject, and a bunch of issues mentioned by Roman, Thanks, Igor On Mon, Oct 19, 2015 at 8:05 PM, Aleksey Kasatkin wrote: > Hi! > > I mostly agree with Igor, GET request should not produce changes (i.e. > allocate VIPs). > >> VIP allocation should be performed when we run deployment. > > I want to give an explanation here. We have a ticket for 8.0 to give an > ability to set arbitrary addresses for VIPs. So, at least some of VIPs can > be set via API calls. And auto-allocation of addresses for VIPs should be > done just before deployment. The only concern here, whether we need a VIP > for net-checker. We could allocate VIPs on net-checker request then also (it > is PUT request). > > AFAIC, we also need to provide an ability to run CheckBeforeDeployment task > separately (only within ApplyChanges for now). It will help the user to > diagnose different problems. It seems to be a subject for another > discussion/ticket though. > > Thanks, > > > > > Aleksey Kasatkin > > > On Mon, Oct 19, 2015 at 11:28 AM, Igor Kalnitsky > wrote: >> >> Hi Roman, >> >> > Not assign addresses to VIPs is a network configuration is being >> > serialized for API output. >> >> AFAIK, that's not truth. Fuel UI and OSTF relies only on *public* VIP. >> So we can keep only *public* VIP, and do not assign / show others. >> >> > Check number of IP addresses wherever it is possible to "spoil" network >> > configuration: when a role get’s assigned to a node, when network >> > changes or network templates are applied. >> >> It won't work that way. What if you enable plugin when all roles are >> assigned? What if you change networks, and now it's not enough IPs? Or >> what if enable plugin that requires 5 VIPs in public network by >> default, and it's not enough. But by using network templates you >> assign this netrole to management network? >> >> From what I can say the proposed approach requires to put checks >> here-and-there around the code. Let's do not overcomplicate things >> without real need. >> >> So let me share my thoughts regarding this issue. >> >> * We shouldn't *allocate* VIPs when we make GET request on network >> configuration handler. It should return only *already allocated* VIPs >> and no more. >> * VIP allocation should be performed when we run deployment. >> * Before deployment checks should fail, if there's not enough VIPs or >> other resources. So users fix them, and try again. >> * Both Fuel UI and OSTF needs VIPs only when cluster is deployed, and >> it's safe to return allocated VIPs on that stage. >> >> So what do you think guys? >> >> Thanks, >> Igor >> >> On Fri, Oct 16, 2015 at 5:25 PM, Roman Prykhodchenko >> wrote: >> > Hi folks! >> > >> > I’ve been discussing several bugs [1-3] with some folks and noticed that >> > they share the same root cause which is that network serialization fails, >> > if >> > there’s not enough IP addresses in all available ranges of one of the >> > available networks to assign them to all VIPs. There are several possible >> > solutions for this issue: >> > >> > a. Not assign addresses to VIPs is a network configuration is being >> > serialized for API output. >> > A lot of external tools and modules, i. e., OSTF, rely on that >> > information so this relatively small change in Nailgun will require big >> > cross-components changes. Therefore this change can only be done as a >> > feature but it seems to be the way this issue must be fixed. >> > >> > b. Leave some VIPs without IP addresses >> > If network configuration is generated for API output it is possible to >> > leave some VIPs without IP addresses assigned. This will only create more >> > mess around Nailgun and bring more issues that it will resolve. >> > >> > c. Check number of IP addresses wherever it is possible to "spoil" >> > network configuration: when a role get’s assigned to a node, when network >> > changes or network templates are applied. >> > >> > >> > The proposal is to follow [c] as a fast solution and file a blueprint >> > for [a]. Opinions? >> > >> > >> > 1 https://bugs.launchpad.net/fuel/+bug/1487996 >> > 2 https://bugs.launchpad.net/fuel/+bug/1500394 >> > 3 https://bugs.launchpad.net/fuel/+bug/1504572 >> > >> > >> > - romcheg >> > >> > >> > __ >> > OpenStack Development Mailing List (not for usage questions) >> > Unsubscribe: >> > openstack-dev-requ...@lists.openstack.org?subject:unsubscribe >> > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev >> > >> >> __ >> OpenStack Deve
Re: [openstack-dev] [Fuel] Assigning VIPs on network config serialization
Hi! I mostly agree with Igor, GET request should not produce changes (i.e. allocate VIPs). > VIP allocation should be performed when we run deployment. I want to give an explanation here. We have a ticket for 8.0 to give an ability to set arbitrary addresses for VIPs. So, at least some of VIPs can be set via API calls. And auto-allocation of addresses for VIPs should be done just before deployment. The only concern here, whether we need a VIP for net-checker. We could allocate VIPs on net-checker request then also (it is PUT request). AFAIC, we also need to provide an ability to run CheckBeforeDeployment task separately (only within ApplyChanges for now). It will help the user to diagnose different problems. It seems to be a subject for another discussion/ticket though. Thanks, Aleksey Kasatkin On Mon, Oct 19, 2015 at 11:28 AM, Igor Kalnitsky wrote: > Hi Roman, > > > Not assign addresses to VIPs is a network configuration is being > > serialized for API output. > > AFAIK, that's not truth. Fuel UI and OSTF relies only on *public* VIP. > So we can keep only *public* VIP, and do not assign / show others. > > > Check number of IP addresses wherever it is possible to "spoil" network > > configuration: when a role get’s assigned to a node, when network > > changes or network templates are applied. > > It won't work that way. What if you enable plugin when all roles are > assigned? What if you change networks, and now it's not enough IPs? Or > what if enable plugin that requires 5 VIPs in public network by > default, and it's not enough. But by using network templates you > assign this netrole to management network? > > From what I can say the proposed approach requires to put checks > here-and-there around the code. Let's do not overcomplicate things > without real need. > > So let me share my thoughts regarding this issue. > > * We shouldn't *allocate* VIPs when we make GET request on network > configuration handler. It should return only *already allocated* VIPs > and no more. > * VIP allocation should be performed when we run deployment. > * Before deployment checks should fail, if there's not enough VIPs or > other resources. So users fix them, and try again. > * Both Fuel UI and OSTF needs VIPs only when cluster is deployed, and > it's safe to return allocated VIPs on that stage. > > So what do you think guys? > > Thanks, > Igor > > On Fri, Oct 16, 2015 at 5:25 PM, Roman Prykhodchenko > wrote: > > Hi folks! > > > > I’ve been discussing several bugs [1-3] with some folks and noticed that > they share the same root cause which is that network serialization fails, > if there’s not enough IP addresses in all available ranges of one of the > available networks to assign them to all VIPs. There are several possible > solutions for this issue: > > > > a. Not assign addresses to VIPs is a network configuration is being > serialized for API output. > > A lot of external tools and modules, i. e., OSTF, rely on that > information so this relatively small change in Nailgun will require big > cross-components changes. Therefore this change can only be done as a > feature but it seems to be the way this issue must be fixed. > > > > b. Leave some VIPs without IP addresses > > If network configuration is generated for API output it is possible to > leave some VIPs without IP addresses assigned. This will only create more > mess around Nailgun and bring more issues that it will resolve. > > > > c. Check number of IP addresses wherever it is possible to "spoil" > network configuration: when a role get’s assigned to a node, when network > changes or network templates are applied. > > > > > > The proposal is to follow [c] as a fast solution and file a blueprint > for [a]. Opinions? > > > > > > 1 https://bugs.launchpad.net/fuel/+bug/1487996 > > 2 https://bugs.launchpad.net/fuel/+bug/1500394 > > 3 https://bugs.launchpad.net/fuel/+bug/1504572 > > > > > > - romcheg > > > > > __ > > OpenStack Development Mailing List (not for usage questions) > > Unsubscribe: > openstack-dev-requ...@lists.openstack.org?subject:unsubscribe > > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev > > > > __ > OpenStack Development Mailing List (not for usage questions) > Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev > __ OpenStack Development Mailing List (not for usage questions) Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
Re: [openstack-dev] [Fuel] Assigning VIPs on network config serialization
Hi Roman, > Not assign addresses to VIPs is a network configuration is being > serialized for API output. AFAIK, that's not truth. Fuel UI and OSTF relies only on *public* VIP. So we can keep only *public* VIP, and do not assign / show others. > Check number of IP addresses wherever it is possible to "spoil" network > configuration: when a role get’s assigned to a node, when network > changes or network templates are applied. It won't work that way. What if you enable plugin when all roles are assigned? What if you change networks, and now it's not enough IPs? Or what if enable plugin that requires 5 VIPs in public network by default, and it's not enough. But by using network templates you assign this netrole to management network? From what I can say the proposed approach requires to put checks here-and-there around the code. Let's do not overcomplicate things without real need. So let me share my thoughts regarding this issue. * We shouldn't *allocate* VIPs when we make GET request on network configuration handler. It should return only *already allocated* VIPs and no more. * VIP allocation should be performed when we run deployment. * Before deployment checks should fail, if there's not enough VIPs or other resources. So users fix them, and try again. * Both Fuel UI and OSTF needs VIPs only when cluster is deployed, and it's safe to return allocated VIPs on that stage. So what do you think guys? Thanks, Igor On Fri, Oct 16, 2015 at 5:25 PM, Roman Prykhodchenko wrote: > Hi folks! > > I’ve been discussing several bugs [1-3] with some folks and noticed that they > share the same root cause which is that network serialization fails, if > there’s not enough IP addresses in all available ranges of one of the > available networks to assign them to all VIPs. There are several possible > solutions for this issue: > > a. Not assign addresses to VIPs is a network configuration is being > serialized for API output. > A lot of external tools and modules, i. e., OSTF, rely on that information so > this relatively small change in Nailgun will require big cross-components > changes. Therefore this change can only be done as a feature but it seems to > be the way this issue must be fixed. > > b. Leave some VIPs without IP addresses > If network configuration is generated for API output it is possible to leave > some VIPs without IP addresses assigned. This will only create more mess > around Nailgun and bring more issues that it will resolve. > > c. Check number of IP addresses wherever it is possible to "spoil" network > configuration: when a role get’s assigned to a node, when network changes or > network templates are applied. > > > The proposal is to follow [c] as a fast solution and file a blueprint for > [a]. Opinions? > > > 1 https://bugs.launchpad.net/fuel/+bug/1487996 > 2 https://bugs.launchpad.net/fuel/+bug/1500394 > 3 https://bugs.launchpad.net/fuel/+bug/1504572 > > > - romcheg > > __ > OpenStack Development Mailing List (not for usage questions) > Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev > __ OpenStack Development Mailing List (not for usage questions) Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
