summaryrefslogtreecommitdiffstats
path: root/redfish-core/lib/ethernet.hpp
Commit message (Collapse)AuthorAgeFilesLines
* Redfish: Implement DHCPv4 and DHCPv6 handlingJohnathan Mantey2019-11-061-95/+256
| | | | | | | | | | | | | | | | | | The Redfish DHCP handling has independent control of DHCPv4 and DHCPv6. This change combines with phosphor-network and phosphor-dbus-interface changes to implement the independent control capability. Tested by: Verified DHCP is able to be enabled for both DHCPv4 and DHCPv6 Verified DHCPv4 can be enabled, and static IPv6 addresses configured Verified DHCPv6 can be enabled, and static IPv4 addresses configured Verified DHCP can be disabled for both interfaces Confirmed enable/disable of UseNTP, UseDNS, and UseHostname Passes Redfish Service Validator Change-Id: I449ec096a3c41231c9bc9aa1bf67824982525cec Signed-off-by: Johnathan Mantey <johnathanx.mantey@intel.com>
* Fix a bunch of warningsEd Tanous2019-10-101-13/+22
| | | | | | | | | | | | | | | using the list of warnings from here: https://github.com/lefticus/cppbestpractices/blob/e73393f25a85f83fed7399d8b65cb117d00b2231/02-Use_the_Tools_Available.md#L100 Seems like a good place to start, and would improve things a bit type-wise. This patchset attempts to correct all the issues in one shot. Tested: It builds. Will test various subsystems that have been touched Signed-off-by: Ed Tanous <ed.tanous@intel.com> Change-Id: I588c26440e5a97f718a0f0ea74cc84107d53aa1e
* Redfish: Make NIC HW reflect PATCH changesJohnathan Mantey2019-09-131-459/+388
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PATCH commands only changed the state of the DBus database. The PATCH did not actually change the state of the NIC settings. IPMITOOL and "ip addr" both delete the specific hash entry, and then create a new entry using the requested state. This change aligns Redfish with that behavior. The requested element is deleted, and subsequently recreated with the new values. This guarantees the NIC HW is actually updated, and in sync with the DBus database. Tested by: From eth1 modify eth0 state: IPv4 tests: Delete all addresses Create five new addresses Add a new entry to the end of the collection Delete the new entry just added Send a "keep" command that does not modify any state, and "keeps" more entries than are actually in the IPv4Addresses collection Keep all entries, and delete entries that do not exist Delete all entries except self assigned Send a PATCH with an empty IPv4StaticAddresses array IPv6 tests: Delete all entries except SLAAC Create five static entries Modify only the prefix for one entry Modify only the address for one entry Add an additional entry Delete the new entry Keep all entries, plus a few that don't exist Keep all entries, delete a few that don't exist Delete all entries except SLAAC Send a PATCH with an empty IPv6StaticAddresses array Change-Id: Id5f733f795588ba36b5d3ab3b0017a01ee3f2da7 Signed-off-by: Johnathan Mantey <johnathanx.mantey@intel.com>
* Network : Fix the Service Validator FailureManojkiran Eda2019-08-271-1/+1
| | | | | | | | | | | | | | | | | | | | | On a DHCP Enabled System, the StaticNameServer property is not aligned to the redfish schema rules and the validator reports the below error: StaticNameServers: Value of Collection property is null but Collections cannot be null, only their entries. Testedby: Redfish Validator *** /redfish/v1/JsonSchemas/EthernetInterface Type (#JsonSchemaFile.v1_0_2.JsonSchemaFile), GET SUCCESS (time: 1.134773) PASS Note: - Previously missed this failure, as i verified the validator on Statically configured setup Signed-off-by: Manojkiran Eda <manojkiran.eda@gmail.com> Change-Id: I6119e8d5783571ac75084301b49df1f44a558c8b
* Avoid populating StaticNameServers propertyManojkiran Eda2019-08-201-1/+9
| | | | | | | | | | | | | | | | | | Redfish ethernet schema has two properties for nameservers 1.NameServers - Represents nameservers currently in use on Interface 2.StaticNameServers -Represents only statically defined set of DNS As we have only one DBUS property `Nameservers` which is used for both the static and dynamic nameservers, So the idea behind this commit is to not populate the StaticnameServers property when DHCP mode is enabled. Tested By: GET https://${BMC_IP}/redfish/v1/Managers/bmc/EthernetInterfaces/eth0 Signed-off-by: Manojkiran Eda <manojkiran.eda@gmail.com> Change-Id: I64f66bde55ec41dc78befbb659a0cbc425bbc24b
* Redfish(Network): IPV4 DHCP ipaddreses dont show gateway when in DHCP modeRavi Teja2019-08-011-2/+3
| | | | | | | | | Tested by: configuring DHCP, GET https://${BMC_IP}/redfish/v1/Managers/bmc/EthernetInterfaces/eth0 Signed-off-by: Ravi Teja <raviteja28031990@gmail.com> Change-Id: Ic7067db41fbb8327ef55cc7552f9153d1ecfb63a
* Redfish(Network): Add support for IPv4StaticAddressesRavi Teja2019-07-031-42/+97
| | | | | | | | | | | | | | | | | Added GET and PATCH support for IPv4StaticAddresses and modified IPv4Addresses as read-only property Tested by: GET PATCH -D patch.txt -d '{"IPv4StaticAddresses": [{},{},{"Address": "10.7.8.7","SubnetMask": "255.255.0.0","Gateway":"10.7.8.1"}]}' PATCH -D patch.txt -d '{"IPv4StaticAddresses": [{},{"Address": "10.7.7.8","SubnetMask": "255.255.0.0","Gateway":"10.7.7.1"}]}' PATCH -D patch.txt -d '{"IPv4StaticAddresses": [null,{},{"Address": "10.7.7.9","SubnetMask": "255.255.0.0","Gateway":"10.7.7.1"}]}']}' PATCH -D patch.txt -d '{"IPv4StaticAddresses": [{},{"Address": "10.8.8.9"}]}' Tested with validator and no errors. Signed-off-by: Ravi Teja <raviteja28031990@gmail.com> Change-Id: I580787933f1d93d0734e32f71ac9fc80dc108247
* [Redfish] Add FQDN and IPv6Address in NetworkProtocol and EthernetInterfaceJennifer Lee2019-06-131-0/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | FQDN and IPv6Address are mandatory properties in OCP profile. Tested: GET /redfish/v1/Managers/bmc/NetworkProtocol HTTP/1.1 Response: { "@odata.context": "/redfish/v1/$metadata#ManagerNetworkProtocol.ManagerNetworkProtocol", "@odata.id": "/redfish/v1/Managers/bmc/NetworkProtocol", "@odata.type": "#ManagerNetworkProtocol.v1_4_0.ManagerNetworkProtocol", "Description": "Manager Network Service", "FQDN": <host name>.<domain name>, "HTTPS": { "Port": 443, "ProtocolEnabled": true }, "HostName": <host name>, ... } GET /redfish/v1/Managers/bmc/EthernetInterfaces/eth0 HTTP/1.1 Response: { "@odata.context": "/redfish/v1/$metadata#EthernetInterface.EthernetInterface", "@odata.id": "/redfish/v1/Managers/bmc/EthernetInterfaces/eth0", "@odata.type": "#EthernetInterface.v1_4_1.EthernetInterface", ... "FQDN": <host name>.<domain name>, "HostName": <host name>, ... "IPv6Addresses": [], ... } Signed-off-by: Jennifer Lee <jennifer1.lee@intel.com> Change-Id: I0ca8f98523cbcfc935e0ac3bbf93f87762ff183b
* Redfish: Remove "state" response content from PATCH commandsJohnathan Mantey2019-06-131-131/+49
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | A PATCH command should not return any JSON content for the NICs. PATCH commands should only return success or failure responses. Reporting state based upon the input to the command is, in most instances, going to be incorrect in comparison to the actual state of the HW. For example: moving from static addresses to a DHCP assigned address cannot be predicted based upon the input. It also takes several seconds for a DHCP transaction to complete, which creates a significant temporal lag between what was sent and what will be. Tested: Performed the following tests: For IPv4: "IPv4Addresses": [ {}, { "Address": "192.168.20.12", "SubnetMask": "255.255.255.0", "Gateway": "192.168.20.1" } ] For IPv6: "IPv6StaticAddresses": [ { "Address": "2001::5:4:3:4", "PrefixLength": 64 } ] Both tests complete without emitting any JSON content. Change-Id: Ic7eb824eb6d996d85d52a3b7c855e9825f4a0d87 Signed-off-by: Johnathan Mantey <johnathanx.mantey@intel.com>
* Redfish: Remove the IPv4 AddressOrigin read-only elementJohnathan Mantey2019-06-131-88/+2
| | | | | | | | | | | | The AddressOrigin entry in the IPv4Addresses schema is read-only. It is not permitted to PATCH the AddressOrigin. Tested: Added the "AddressOrigin": "DHCP" to an IPv4Addresses PATCH command. The JSON response is "property is not in the list of valid properties". Change-Id: Id7d41e9f1b393738e20725495b3821dd2f9f69d0 Signed-off-by: Johnathan Mantey <johnathanx.mantey@intel.com>
* Redfish: Correctly identify IPv4 LinkLocal vs. LinkGlobalJohnathan Mantey2019-06-131-2/+2
| | | | | | | | | Self assigned IPv4 addresses must be flagged as LinkLocal, not LinkGlobal. Likewise global IPv4 addresses must be flagged as LinkGlobal. Change-Id: I9f6f0ceb3d312dd0198cb54c59c579ed96fbc38a Signed-off-by: Johnathan Mantey <johnathanx.mantey@intel.com>
* Redfish: Correct PATCH for IPv6 static addressesJohnathan Mantey2019-06-101-12/+4
| | | | | | | | | Manipulating IPv6 static addresses works the same way as IPv4 static addresses. Keep the IPv6StaticAddresses collection, and remove write access from the read-only IPv6Addresses collection. Change-Id: If58c858c676f7ae843e0642800a4083c5df79d1d Signed-off-by: Johnathan Mantey <johnathanx.mantey@intel.com>
* Redfish(Network): Add support for IPv6Addresses and IPv6StaticAddressesRavi Teja2019-06-061-60/+445
| | | | | | | | | | | | | | | | | | | Added GET support for IPv6Addresses Added GET and PATCH support for IPv6StaticAddresses Tested by: GET PATCH -D patch.txt -d '{"IPv6StaticAddresses": [{"Address": "2002:905:150e:301:72e2:84ff:fe14:222","PrefixLength": 64}]}' PATCH -D patch.txt -d '{"IPv6StaticAddresses": [{},{"Address": "2002:905:150e:301:72e2:84ff:fe14:333","PrefixLength": 64}]}' PATCH -D patch.txt -d '{"IPv6StaticAddresses": [null,{},{"Address": "2002:905:150e:301:72e2:84ff:fe14:444","PrefixLength": 64}]}' PATCH -D patch.txt -d '{"IPv6StaticAddresses": [{"Address": "2002:905:150e:301:72e2:84ff:fe14:555","PrefixLength": 64},{}]}' PATCH -D patch.txt -d '{"IPv6StaticAddresses": [{},{"Address": "2002:905:150e:301:72e2:84ff:fe14:666"}]}' PATCH -D patch.txt -d '{"IPv6StaticAddresses": [{},{"PrefixLength": 64}]}' Tested with validator and no errors. Change-Id: I7d1314a0c7843aae8425d66119f0d205a5cfac55 Signed-off-by: Ravi Teja <raviteja28031990@gmail.com>
* [Redfish] Enable PATCH for DHCPv4 properties in EthernetInterface schemaJennifer Lee2019-05-291-1/+86
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | DHCPv4 were read-only properties, this code change make following properties to be PATCHable: - DHCPEnabled - UseDNSServers - UseDomainName - UseNTPServers Tested: 1. Check DHCPv4 properties value GET /redfish/v1/Managers/bmc/EthernetInterfaces/<ethID> HTTP/1.1 Response: { ...... "DHCPv4": { "DHCPEnabled": true, "UseDNSServers": true, "UseDomainName": true, "UseNTPServers": true }, ...... } 2. PATCH DHCPv4 PATCH /redfish/v1/Managers/bmc/EthernetInterfaces/eth1 HTTP/1.1 { "DHCPv4": { "DHCPEnabled":false, "UseDNSServers": false, "UseDomainName": false, "UseNTPServers": false } } 3. Check whether DHCPv4 properties are set GET /redfish/v1/Managers/bmc/EthernetInterfaces/<ethID> HTTP/1.1 Response: { ...... "DHCPv4": { "DHCPEnabled": false, "UseDNSServers": false, "UseDomainName": false, "UseNTPServers": false }, ...... } Signed-off-by: Jennifer Lee <jennifer1.lee@intel.com> Change-Id: I125029ae4cdef329edabd18f6aa6a16db6e27d3d
* Fix 404 handling in Redfish EthernetInterfacesEd Tanous2019-05-201-16/+40
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously, when a user requested a network interface that was non-sensical, like: /redfish/v1/Managers/bmc/EthernetInterfaces/foobar/VLANs OR /redfish/v1/Managers/bmc/EthernetInterfaces/foobar They would be presented with a 200-OK, and a partially filled in object. While this doesn't matter much for the casual redfish user, who uses the collection to properly enumerate devices, this causes an issue with security scanners, which think they've found some vulnerability when they can throw whatever injection text they want to in the message, and it shows up in the response. This patchset corrects this, and causes the urls referenced above to properly return 404, and the appropriate "ResourceNotFound" error message. Tested: Attempted both URLs shown above. Both return 404. Ran redfish service validator, observed no errors related to EthernetInterface, or sub nodes. Attempted good URLs, and observed no change to the payload. Signed-off-by: Ed Tanous <ed.tanous@intel.com> Change-Id: Idb2758858d4dbaf421c0cef28b1d5d02402e0ad8
* Redfish(Network): Implemented PATCH of Nameservers arrayRAJESWARAN THILLAIGOVINDAN2019-05-091-2/+11
| | | | | | | | | | | | | | | | | | | | | Testing: "NameServers": [], "NameServers@Message.ExtendedInfo": [ { "@odata.type": "/redfish/v1/$metadata#Message.v1_0_0.Message", "Message": "The property NameServers is a read only property and cannot be assigned a value.", "MessageArgs": [ "NameServers" ], "MessageId": "Base.1.4.0.PropertyNotWritable", "Resolution": "Remove the property from the request body and resubmit the request if the operation failed.", "Severity": "Warning" } ] Change-Id: I12b8cb0ad9b8ad83371ee1d834f38e7537cc7b61 Signed-off-by: RAJESWARAN THILLAIGOVINDAN <rajeswgo@in.ibm.com>
* Redfish(Network): Implement IPv6DefaultGateway supportRavi Teja2019-04-301-1/+22
| | | | | | | | | | | | | | Added GET support for IPv6DefaultGateway.its read only property On Patch operation display "PropertyNotWritable" message. Tested by: GET PATCH -D patch.txt -d '{"IPv6DefaultGateway": "fe80::226:88ff:feac:8401"}' Tested with validator and no errors. Change-Id: Ib67234496b70ec53508a9be36cd86c1859b5d895 Signed-off-by: Ravi Teja <raviteja28031990@gmail.com>
* Fix empty IPV4 gateway errorGunnar Mills2019-04-261-1/+8
| | | | | | | | | | | | | | | If an IPV4 gateway is empty use "0.0.0.0". The RedfishServiceValidator throws the following error on a Witherspoon system: "IPv4Addresses[0].Gateway: String '' does not match pattern ''^(?:[0-9]{1,3}\\.){3}[0-9]{1,3}$''" Tested: Ran RedfishServiceValidator and no longer observe the error. IPV4 addresses with gateways are untouched. Change-Id: I423aeeb0bec46fa5cc0c8cf5a017c16d20314984 Signed-off-by: Gunnar Mills <gmills@us.ibm.com>
* Redfish(Network): Implemented PATCH of StaticNameServers arrayRAJESWARAN THILLAIGOVINDAN2019-04-051-3/+38
| | | | | | | | | | | Testing: PATCH -D patch.txt -d '{"StaticNameServers": ["3.3.3.3"]}' PATCH -D patch.txt -d '{"StaticNameServers": ["3.3.3.3","6.6.6.6"]}' PATCH -D patch.txt -d '{"StaticNameServers": []}' Signed-off-by: RAJESWARAN THILLAIGOVINDAN <rajeswgo@in.ibm.com> Change-Id: I3ffb7f5b704437b31ddfeb6a74eae3e6a582164a
* Redfish(Network): VLAN PATCH fixSunitha Harish2019-04-051-64/+46
| | | | | | | | | | | | | | | This patch set fixes the issue at the Patch operation on VLAN resource. Before this for any new value of VLANId , the patch command used to apply value 1 as the new VLANId Tested by : GET https://${bmc}/redfish/v1/Managers/bmc/EthernetInterfaces/eth0/VLANs/eth0_10 PATCH https://${bmc}/redfish/v1/Managers/bmc/EthernetInterfaces/eth0/VLANs/eth0_10 -d '{ "VLANId" : 12, "VLANEnable": true}' PATCH https://${bmc}/redfish/v1/Managers/bmc/EthernetInterfaces/eth0/VLANs/eth0_10 -d '{ "VLANId" : 20, "VLANEnable": false}' Signed-off-by: Sunitha Harish <sunithaharish04@gmail.com> Change-Id: Ia18e58816e6c78c8609e0daa558bc66d6914568e
* Redfish(Network): Fixes VLAN specific issuesSunitha Harish2019-03-281-97/+60
| | | | | | | | | | | | | | | | | | | | This commit fixes the following issues. Before this fix the VLAN functionality was inconsiststant. 1)Add VLANs property to the EthernetInterface resource 2)Removed VLAN property under EthernetInterface resource 3)Made the property - VLANEnable mandatory for creating vlans (POST command) 4)Fetch the VLANId when GET on the VLAN Interface resource 5)Updated Ethernet schema to EthernetInterface.v1_4_1 6)Changed the propert name "DHCPv4Configuration" to "DHCPv4" as per the schema Tested by: GET https://${bmc}/redfish/v1/Managers/bmc/EthernetInterfaces/eth0 GET https://${bmc}/redfish/v1/Managers/bmc/EthernetInterfaces/eth0/VLANs POST -D headers.txt https://${bmc}/redfish/v1/Managers/bmc/EthernetInterfaces/eth0/VLANs -d '{ "VLANId" : 30, "VLANEnable": true}' GET https://${bmc}/redfish/v1/Managers/bmc/EthernetInterfaces/eth0/VLANs/eth0_30 Signed-off-by: Sunitha Harish <sunithaharish04@gmail.com> Change-Id: I2339f6711cdd56fe42fee030701125492911dc26
* Revert "bmcweb: Fix a bunch of warnings"Ed Tanous2019-03-251-9/+47
| | | | | | | | This reverts commit 6ea007a2faec52ad62680015d2a3f00371a1e351. Reason for revert: Reports of bmcweb seg faults. Change-Id: I408f1bb29c2f8e427a6621cdaac8c31b847ebf06
* bmcweb: Fix a bunch of warningsEd Tanous2019-03-221-47/+9
| | | | | | | | | | | | bmcweb classically has not taken a strong opinion on warnings. With this commit, that policy is changing, and bmcweb will invoke the best warnings we are able to enable, and turn on -Werror for all builds. This is intended to reduce the likelihood of hard-to-debug situations that the compiler coulve caught early on. Change-Id: I57474410821e82666b3a108cfd0db7d070e8900a Signed-off-by: Ed Tanous <ed@tanous.net>
* Redfish(Ethernet): Send the correct JSON response after creating the IPV4 ↵Ratan Gupta2019-03-191-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | object. Before this commit after creating the IPv4 address object,JSON response was having empty string for the newly added object. This commit fixes the problem. Tested By: Assuming there are already two IP addresses configured then the following PATCH request would throw the following output PATCH -D patch.txt -d '{"IPv4Addresses": [{},{},{"Address": "a.a.a.a","Address Origin": "Static","SubnetMask": "255.255.0.0","Gateway":"g.g.g.g"}]} { "@odata.id": "/redfish/v1/Managers/bmc/EthernetInterfaces/eth0", "HostName": "witherspoon", "IPv4Addresses": [ { "Address": "169.254.201.17", "AddressOrigin": "IPv4LinkLocal", "Gateway": "9.3.185.1", "SubnetMask": "255.255.0.0" }, { "Address": "9.3.185.83", "AddressOrigin": "Static", "Gateway": "9.3.185.1", "SubnetMask": "255.255.255.0" }, { "Address": "a.a.a.a", "Gateway": "g.g.g.g", "SubnetMask": "255.255.0.0" } ], Change-Id: I6be340051020839ff7f669296c2077ecfdce78ff Signed-off-by: Ratan Gupta <ratagupt@linux.vnet.ibm.com>
* Redfish(Network): Support the PATCH opertration of MACAddress propertyRatan Gupta2019-03-191-2/+29
| | | | | | | | TestedBy: PATCH '{"MACAddress": "x2:xx:xx:xx:xx:xx"}' Change-Id: I6dc421ea509506a2f3062fc0531778ffa2b00310 Signed-off-by: Ratan Gupta <ratagupt@linux.vnet.ibm.com>
* Redfish(Network): Support for deletion of IP address through PATCH operationRatan Gupta2019-03-191-76/+80
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Before this commit we were reading the IPV4address through readJSON function in vector of json, readJSON doesn't like the null list parameter and throws the error as below "The value null for the property IPv4Addresses is of a different type than the property can accept" This commit fixes that problem and some restructuring the code to make delete work. TestedBy: PATCH {"IPv4Addresses": [{},null,{},{}]} deletes second object PATCH {"IPv4Addresses": [null,{}]} deletes first object Suppose if the IPv4Address has two addresses then the Following PATCH request would give the folowing error. PATCH {"IPv4Addresses": [{},{},null] { "IPv4Addresses/0@Message.ExtendedInfo": [ { "@odata.type": "/redfish/v1/$metadata#Message.v1_0_0.Message", "Message": "The value [{},{},null] for the property IPv4Addresses/2 is of a different format than the property can accept.", "MessageArgs": [ "[{},{},null]", "IPv4Addresses/2" ], } Change-Id: Iaefd59de469cb345e86b19349b27567a4fcec3fa Signed-off-by: Ratan Gupta <ratagupt@linux.vnet.ibm.com>
* Redfish: Populate DHCPv4Configuration Informationmanojkiraneda2019-03-181-0/+58
| | | | | | | | | | | | | | | | | | With this commit the get-request on ethernet interface service gets the DHCPv4Configuration.The following are the properties that are add as a part of this commit. - UseDNSServers - UseNTPServers - UseDomainName - DHCPEnabled Tested By:(On Qemu) GET https://<ip>/redfish/v1/Managers/bmc/EthernetInterfaces/eth0 Change-Id: Ia1f6dbe2c4f1f2ed33adc54f7fd99b61a36c1058 Signed-off-by: manojkiraneda <manojkiran.eda@gmail.com>
* Redfish(Network): Allow empty list item for ipv4 address.Ratan Gupta2019-03-121-0/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Tested by: Assuming there are two IP addresses in the IPv4Addresses for the following PATCH request. 1) PATCH {"IPv4Addresses": [{},{}]} =>No change in the existing list. 2) PATCH {"IPv4Addresses": [{},{},{}]} Following error for the third list item. "IPv4Addresses/0/Address@Message.ExtendedInfo": [ { "@odata.type": "/redfish/v1/$metadata#Message.v1_0_0.Message", "Message": "The property IPv4Addresses/2/Address is a required property and must be included in the request.", "MessageArgs": [ "IPv4Addresses/2/Address" ], "MessageId": "Base.1.4.0.PropertyMissing", "Resolution": "Ensure that the property is in the request body and has a valid value and resubmit the request if the operation failed.", "Severity": "Warning" } Change-Id: I24d11ca82cf6843611f72912499878bcbe1aecac Signed-off-by: Ratan Gupta <ratagupt@linux.vnet.ibm.com>
* Redfish(Ethernet): Fix to not convert the json index into stringRatan Gupta2019-03-121-6/+5
| | | | | | | | | | | | | | | | | During Property Update for the Ipv4address, Index was getting converted into string which was causing the bmcweb to crash. This commit fixes this problem. TestedBy: Assumption: There is already three IP address and updating the third one. PATCH -D patch.txt -d '{"IPv4Addresses": [{},{},{"Address": "x.x.x.x","AddressOrigin": "Static","SubnetMask": "255.255.0.0","Gateway":"x.x.x.x"}]} Change-Id: Ic055649b75aaa36e16bda5d8dc4e4e9c87dcb315 Signed-off-by: Ratan Gupta <ratagupt@linux.vnet.ibm.com>
* Redfish(Ethernet): Don't init the vlanID and vlanEnableRatan Gupta2019-03-121-2/+4
| | | | | | | | | | | | | | | | | | These variables are of type std::optional,The values should be filled by the readJSON function,as further decision to handle the patch for the vlan is dependent on whether the value for vlanID and vlanEnable is not null. Due to this issue, control goes to handle the vlan patch and the readJSON inside the handleVlanPatch doesn't find the values for vlanID and vlanEnable so set the resp result to forbidden. After that if any other readJSON would be called it would return failure and due to that no other PATCH operation on the ipv4Address can be completed. Change-Id: I3ebab18ae4adfa7c871ecab862a2865dc14f154d Signed-off-by: Ratan Gupta <ratagupt@linux.vnet.ibm.com>
* bmcweb: move ethernet VLAN to readJsonEd Tanous2019-02-271-124/+64
| | | | | | | | This commit moves the various vlan interfaces to use the existing std::vector based implementations, and not the existing implementations. Change-Id: Id2503d5e2f1503b61ec7dbdb25098611382b24c5 Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* bmcweb: fix compiler warningsEd Tanous2019-02-211-4/+2
| | | | | | | | | | | | This patchset attempts to fix all compiler warnings in bmcweb owned files. There are 2 warnings left, both in sdbusplus, which will be resolved in a patchset there. Tested By: Recompiled, observed warning count went from 30, to zero. Change-Id: Ife90207aa5773bc28faa8b04c732cafa5a56e4e4 Signed-off-by: Ed Tanous <ed@tanous.net>
* bmcweb: Implement single Redfish "system" endpointEd Tanous2019-02-191-34/+61
| | | | | | | | | | | | | | | | | | | | | | This commit changes the redfish behavior to move to a single, known name under the /redfish/v1/Systems/system path. This is advantageous for a lot of reasons. 1. Lots fewer dbus calls to determine the system name for every path. This could be optimized in other ways, like checking the system name on startup, but because redfish paths are not intended to be informative, this patchset takes the opinion that less code is better. 2. Lots of lowered complexity, given that each endpoint underneath /system doesn't need an individual "does this system exist, and is the name right" check. 3. This makes it possible to correctly implement the "Links" property in Chassis, which is required for the OCP base server profile Tested By: Very minimal testing done, but it seems to pass the validator. Change-Id: Iea3cb5081b92a3843b6877decd009936de00561c Signed-off-by: Ed Tanous <ed.tanous@intel.com> Signed-off-by: Gunnar Mills <gmills@us.ibm.com>
* bmcweb: move variant usage to std namespaceEd Tanous2019-02-091-36/+24
| | | | | Change-Id: I9d7069668f91f2ac72d2f4a440f63e0e85dd5269 Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* Use the unpacking mechanisms for hostnameEd Tanous2019-01-041-42/+15
| | | | | | | | | The ethernet interface previous hardcoded error handling patterns for the interfaces shown. This commit moves it to the generic ReadJson mechanisms for error handling Change-Id: I710188c808ee62c5797059d7ee8f9876bf872a11 Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* bmcweb: Enable redfish unpacking of complex typesEd Tanous2018-12-181-128/+72
| | | | | | | | | | | In certain cases, redfish wants to unpack specialized types that might be more complex than just simple structs. This commit adds the ability to unpack a nlohman json object directly, and moves the ethernet schema over to the new api Change-Id: Ib3c25e6c4d8f0a163647092adfb454355d329170 Signed-off-by: Ed Tanous <ed.tanous@intel.com> Signed-off-by: James Feist <james.feist@linux.intel.com>
* bmcweb: Move boost::optional to std::optionalEd Tanous2018-12-101-3/+3
| | | | | | | | This change is largely a sed replace, and gets us closer to standards compliance. Change-Id: Ie60437eff43de830ef19389344a55430dabd9d16 Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* Remove custom version of getPtrEd Tanous2018-12-101-19/+22
| | | | | | | | Now that sdbusplus variant supports std::get_if, we can remove our custom, mapbox namespaced implementation that does the same thing. Change-Id: I854c473003e28e41dd45dba08ca683433f1c1774 Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* bmcweb: Redfish away from json cacheEd Tanous2018-11-261-66/+53
| | | | | | | | | | | | | | | | | | | | In the original incarnation of bmcweb, route registration was done automatically. This has proved to be a terrible idea, wraught with corner cases and issues. The route registration is currently the only user of the redfish::Node::json element. Unfortunately, as written, this structure consumes a lot of memory that's duplicated and not very useful. From a performance perspective, there is almost no difference between rebuilding the structure for each GET request, and having the "cache" that needs to be copied into the response and modified before it can be useful. In the programming tradeoffs for bmc, lower memory usage is more important than latency, especially at these levels. Change-Id: I785e8352123e5e886acf05cd59cb23648f93839d Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* Simplify Redfish error message property interfaceJason M. Bills2018-11-151-78/+51
| | | | | | | | | | | | | | | The error message code used a json_pointer object which must begin with a '/' character and had to be sent as an extra parameter. This change simplifies the interface by using a string so there doesn't have to be a '/'. This allowed the same property argument passed for the message to be used for the property field path. Tested: Sent an error with a property and verified that it is correctly displayed. Change-Id: I0571e2eee627cedf29d751588a4f1bceee66f084 Signed-off-by: Jason M. Bills <jason.m.bills@linux.intel.com>
* Improve the Redfish error reporting interfaceJason M. Bills2018-10-221-285/+189
| | | | | | | | | | | Makes the Redfish error reporting interface automatically handle setting the http status and JSON content in the response object. When using an AsyncResp object, this allows for simply calling the Redfish error and returning. Change-Id: Icfdce2de763225f070e8dd61e591f296703f46bb Signed-off-by: Jason M. Bills <jason.m.bills@linux.intel.com>
* Cleanups to ethernet schemasEd Tanous2018-10-161-1391/+1225
| | | | | | | | | | | | | | Several rounds of cleanup to the ethernet interfaces. Changes involved: 1. Move schemas to std::shared_ptr for handling responses 2. Remove the use of the "providers" interface. As there is currently only one usage, the providers interface doesn't give us any flexibility. 3 Move the ethernet interfaces to flat_set, to allow better sorting and lookup times. 4. Fix a number of "use after free" bugs with lambda callbacks that were passing by reference instead of by value. Change-Id: I3f8ffb10702563b25fda88155f1929b8d8e38bf5 Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* Implement redfish LogServiceEd Tanous2018-10-091-0/+1
| | | | | | | | | | | This commit implements an optional LogService endpoint for debugging CPUs over Redish. It exposes both the ability to create a new log with the LogImmediate action, as well as to pull a stored log using the LogCollection schema. This feature is disabled by default, and should be enabled in the meta-x86 layer. Change-Id: Ied1d5e263a0857e09cd9a24fddec9c11a4066cd2 Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* Move to clang-format-6.0Ed Tanous2018-09-051-1568/+1880
| | | | | | | | This commit moves the codebase to the lastest clang-format file from upstream, as well as clang-format-6.0. Change-Id: Ice8313468097c0c42317fbb9e10ddf036e8cff4c Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* Last round of variable renamesEd Tanous2018-08-151-10/+9
| | | | | | | | A few patches were in flight when we did the great variable renaming. This catches all the patches that have been merged since Change-Id: Ie9642a4812b2a679ffa01ce540f5c4515a4c710b Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* Clang-format cleanupEd Tanous2018-08-131-147/+146
| | | | | | | | | | | | | | | | | | | | | | It looks like some of the code reviews in progress contain clang-format fixes as well. This squashes all of the clang-format fixes into a single commit, that hopefully can go through the process a little faster, as it should only be whitespace changes. In the past, clang-formatting the naughty strings file proved to return inconsistent results, and even formatted files would still be detected as unformatted. Aparently the solution was to run clang-format-5.0 approximately 20 times in a loop, until it decided on a code layout that it was ok with iteration to iteration. This is a bit of a corner case, so I don't think we need to get CI running multiple runs in the future, and can just handle this on a case by case basis. I'd be surprised if anyone else had this issue. Change-Id: I57e2a03676bce20dc376fd9cea724732b2dc7010 Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* Fix merge conflictEd Tanous2018-08-091-178/+171
| | | | | | | | | Got a couple patches that collided in air, and now builds are broken. This resolves the collision by moving the new patches forward to the latest #defines Change-Id: I1fe35d17a68c61ad90752ae73000e2579131bf5d Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* Move over to upstream c++ styleEd Tanous2018-07-261-310/+306
| | | | | | | | | | | | | | | | | | This patchset moves bmcweb over to the upstream style naming conventions for variables, classes, and functions, as well as imposes the latest clang-format file. This changeset was mostly built automatically by the included .clang-tidy file, which has the ability to autoformat and auto rename variables. At some point in the future I would like to see this in greater use, but for now, we will impose it on bmcweb, and see how it goes. Tested: Code still compiles, and appears to run, although other issues are possible and likely. Change-Id: If422a2e36df924e897736b3feffa89f411d9dac1 Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* Make Ethernet/VLAN Interfaces handle "404 Not Found" properlyKowalski, Kamil2018-07-261-6/+134
| | | | | | | | | | | | | | Ethernet and VLAN Interfaces lacked detection whether interface actually existed. So this detection had to be added. Additionally there was slight inconsistency with error messages and statuses, this is fixed now. Tested on x86 development machine and dedicated platform. Does not introduce any regression. Change-Id: Ic9e2a98c748ac0dd1cf412b6a54a14f3f53eb9fa Signed-off-by: Kowalski, Kamil <kamil.kowalski@intel.com> Signed-off-by: Ed Tanous <ed.tanous@intel.com>
* Make SessionStore a proper singletonBorawski.Lukasz2018-07-251-2/+0
| | | | | | | | | | | | | | | - SessionStore class now has a proper singleton structure - session_storage_singleton.hpp is removed - from_json(..) function for SessionStore is changed to a specialized template - minor cosmetic fixes added - Move the template class usages of Crow App over to a non-template parameter Change-Id: Ic9effd5b7bac089a84c80a0caa97bd46d4984416 Signed-off-by: Borawski.Lukasz <lukasz.borawski@intel.com> Signed-off-by: Ed Tanous <ed.tanous@intel.com>
OpenPOWER on IntegriCloud