[Device]: Huawei OceanProtect E8000 Backup Storage #742
Labels
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
jaylouisw/packetforge#742
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Vendor
Huawei
Exact model name
Huawei OceanProtect E8000 Backup Storage
Category
NAS
Ports and interfaces
PoE budget (watts)
None
Datasheet or product page URL
https://e.huawei.com/en/products/storage/oceanprotect/oceanprotect-e8000
What do you want to model with it?
Backup storage for medium/large enterprises (high capacity + reduction).
Capabilities
backup storage (NFS/SMB/iSCSI/S3/FC target), dedupe/compression
Before submitting
Blocked on a port count, not on research effort.
Huawei's public page for the OceanProtect E8000 gives the port types - front-end 10/25/40/100GE and 16/32/64 Gbit/s Fibre Channel, with 2 or 4 controllers per node and up to 8 in a system - but not how many front-end ports a node has. That number depends on the interface-card configuration, and the configuration-specific tables live on
support.huawei.com, which is behind a login wall I cannot pass (plain fetch returns HTTP 422; a real browser returns an empty page).A storage array with no stated port count would have to be invented, so this stays open. If someone can paste the front-end port count for the requested configuration, I will land it the same day.
Closing status: blocked on a stated configuration.
Huawei publishes OceanProtect front-end port types - "16/32/64 Gbit/s FC; 10/25/40/100GE" - and a controller count, but not a port count, because how many front-end ports a backup array has is a function of the controllers and interface cards ordered. E8000's page publishes no interface table at all.
The X-series comparison table is otherwise complete (capacity, bandwidth, controllers, cache), which is why this is a narrow blocker rather than an unreadable page.
What would close it: a controller/interface-card configuration, or an authoritative Huawei ordering guide that pins ports per base model.
docs/DEVICE_REGISTRY.md§"What is left open, and why" records the same reason for all three (#741, #742, #743).Re-checked at the datasheet level. Still blocked on the same field - and now provably unpublished, not merely unfound: the E8000 has a primary datasheet, it was read end to end today, and it stops at port types like the marketing page.
Sources read (2026-11-17)
https://e.huawei.com/en/products/storage/oceanprotect/oceanprotect-e8000- 200, 343,723 bytes.https://e.huawei.com/en/documents/products/storage/597eb7d1363f4b5da0d4ad663da1f7b6- the datasheet wrapper, 200..../marketingcloud/pep/asset/20000001/Material/597eb7d1363f4b5da0d4ad663da1f7b6/M3T1A590N1325408746781700176/Huawei%20OceanProtect%20E8000%20Backup%20and%20Archiving%20Converged%20Storage%20V200R001C20%20-%20Data%20Sheet.pptx- 200, 2,407,538 bytes,application/vnd.openxmlformats-officedocument.presentationml.presentation. The datasheet is a PowerPoint, which is why three passes of PDF-hunting never found it.https://support.huawei.com/hedex/api/navi/EDOC1100486186/first-level-tree-401 {"status":401,"message":"NOT LOGIN!"}.https://support.huawei.com/enterprise/en/doc/EDOC1100318730/ce945100/cabinet-layout-and-connection-planning- 200, 567,916 bytes (method control).What slide 3 of the datasheet says about ports
There is no port-count row, no interface-module table and no slot count. The page's "2 U 36-slot high-density hardware" is the disk slot count and must not be read as a port census.
Also worth flagging: the request quotes 6.2 / 19 / 55 TB/hour and 8 nodes / 16 PB. Those are the X-series tier table, not the E8000. The page says 53 / 115 / 55 TB/hour and 128 nodes / 256 PB; the datasheet says "up to 55" physical, "up to 110" logical, "up to 46 (all-flash) / 15 (hybrid flash)" recovery per node, and it lists NL-SAS disks (8/10/16/20/24 TB) that the page omits. The two Huawei documents disagree on bandwidth; neither is a port census, so the discrepancy is recorded rather than resolved.
Exact missing fields: interface modules per controller, which of the controller's slots accept front-end modules, and each module's port count.
What would unblock it: any indexed section of
EDOC1100486186(Product Description) orEDOC1100486185(Installation Guide) - one section opens the whole set, because the SSR pages embed the document's full TOC (31 section links on the control page above); or a Huawei ordering guide pinning ports per base model; or one stated configuration in the shape "E8000 node, 2 controllers, 2 x 4-port 25GE interface module per controller".One trap recorded so the next pass does not fall into it: the accessible X6000/X8000 1.3.0 installation guide says "each of controllers A and B provides six interface module slots ... front-end modules only in IOM 0, IOM 1 and IOM 4". The E8000 is a different enclosure (86.1 x 447 x 820 mm NVMe controller enclosure), and no Huawei document read says the two share a controller - so those six slots do not transfer.
Research note:
research/oceanprotect-x-series-port-census.md, PR #1785. That PR records the evidence; it does not close this issue.Still blocked - and this pass closed the illusion that a nearby table answers it. PR #1833 records the evidence in
research/oceanprotect-x-series-port-census.md�8-10; the issue stays open.The first search result for "OceanProtect interface module port count" is Dorado.
https://info.support.huawei.com/hedex/api/pages/EDOC1100214756/YEK0928G/12/resources/plan_plan_all.html- Connection Planning Between the Storage System and Application Server - prints exactly the table this issue needs:GE electrical 4 ports | 10GE electrical 4 | 40GE 2 | 100GE 2 | 25G RoCE 4 | 100G RoCE 2 | 8/16/32/64G FC 4 | 10/25GE 4 | 25G NoF 4 | 100G NoF 2. It is OceanStor Dorado V6: the same EDOC'sdisk_encryt_general.htmltopic reads "This document describes the disk encryption feature of OceanStor Dorado storage systems", its other topics aredorado_18000_021/091, and its attachmenten-us_attachment_0000001207425401.pdfnames "OceanStor Dorado V6 series ... Dorado 18000 V6". Second instance of the Dorado Table 6-1 hazard already noted for this family. Fetching those topics now answers 200 with the 91,294-byte support index, so only the search index holds that text - and it belongs to another product.The E8000 set, identified and still shut.
https://support.huawei.com/enterprise/en/data-protection/oceanprotect-storage-x-series-pid-262802350(200, 472,195 B) publishes all three books with the idPath the site itself uses: Product DescriptionEDOC1100486186, Quick Installation GuideEDOC1100485347(id not previously recorded here), Hardware Installation GuideEDOC1100486185. Every one of them, andEDOC1100486186?lang=enand the zh-locale root, answers 302 tohttps://support.huawei.com/view/notFound- a 452,479-byte shell with no document links.idPathis navigation context, not a resolver.One candidate book eliminated.
Huawei OceanProtect X Series Backup Appliance Technical White Paper, Issue 01 (2026-09-29), 167 pp, 16,574,915 B, resolved fromhttps://e.huawei.com/en/documents/products/storage/eded1b265ecc4dfba2fbf8669b0b8213:E8000occurs zero times (X6000 24, X8000 27, X8100 11, X9000 29, X9100 22). Its "36 (palm-sized) NVMe disk slots" line is the X8100 enclosure, not the E8000's 2 U 36-slot body.Exact missing fields: interface-module slots per controller on the E8000 node, and the port count of each front-end module type (16/32/64 Gb FC; 10/25/40/100GE).
What would unblock it, any one of: a section URL of
EDOC1100486186/EDOC1100485347/EDOC1100486185(one is enough - the fetchable viewer then exposes the whole TOC), an E8000 order line such as "E8000 node, 2 controllers, 2 x 4-port 25GE IOM per controller", orshow interface_moduleoutput from a delivered array. No spec will be written from Dorado's or X8100's numbers.Still blocked on the same field - but both remaining routes in this thread are now measured, not assumed. PR #1873 carries it (
research/oceanprotect-e8000-enumeration-gates.md); the issue stays open.Route 1 - the document tree: a Huawei account, at the service, not at the edge
The section URLs that do render all carry a revision-scoped topic hash, and the tree that holds them is
GET https://support.huawei.com/supportgateway/view/v1/enterprise/doc/catalogue?nid=EDOC1100486186-> HTTP 401
Control on the same URL: a browser user-agent gets Akamai
403 Access Denied, a crawler user-agent gets the 401 JSON. So the 403 is the edge refusing the client and the 401 is the document service refusing the request - the endpoint is reachable and wants a session. One reading trap worth recording: the 401 body is gzip-encoded whateverAccept-Encodingsays, so withoutcurl --compressedit reads as an empty 91-byte response, which is how an earlier pass could conclude "no answer".That is the mechanism behind a fact this thread already had: an E8000 book renders only when a topic hash is already indexed by a search engine. The X3000 V200R001 Product Description renders (
EDOC1100368234/8587c2ff/overview);EDOC1100486186has no indexed hash, and there is no anonymous way to ask for the list.Route 2 - Spare Parts Query: public CSRF handshake, private data
https://info.support.huawei.com/storage/spareparts/is a public 200 (bundlemain-ZCQDDWQH.js, 2,605,430 B) and names its own call:{API_HOST}/storage-gw/rest/itcms/getDataList?appId=spare&dataSetId=SpareParts_Product_Output&nodeId=SpareParts_Product_Output&lang=en&child=true.GET /storageinfogw/csrf-gw/rest/csrf/getCsrfTokenGET .../getDataList?appId=spare...withCsrf-TokenX-Csrf-Token,csrf-tokenReferer, withlang: enGET /storageinfo-gw-v2/csrf-gw/rest/csrf/getCsrfTokenA token issued to anyone and then not honoured on the data call is a service checking a session of its own.
https://info.support.huawei.com/storage/spec/uses the sameAPI_HOST/API_HOST_V2pair, so Specifications Query - the route recorded for #743 - is behind the same gate.What this does not change
The only anonymously published E8000 document is still the V200R001C20 data sheet: re-fetched today through
scripts/huawei-material-fetch.js, 2,407,538 B, 4 slides, withslot0,interface module0,IOM0 andport1 occurrence over the whole deck.Exact missing fields, unchanged: the interface-module slots per controller on the E8000 node, and the port count of each front-end module type (10/25/40/100GE, 16/32/64 Gbit/s FC) as fitted to it.
What would close it, any one of:
EDOC1100486186(Product Description),EDOC1100485347(Quick Installation Guide - Huawei's own product page now lists it, dated 2026-09-30) orEDOC1100486185(Installation Guide), pasted by a reader with a support session;No spec is written from the X6000/X8000 enclosure: equal 820 mm depth is not a shared backplane, and the one document that names all three (EDOC1100305200, Parts Replacement) only makes the 2 U fan module and smart disk enclosure shared FRUs. Six borrowed IOM slots on that basis would be the family-sibling error this catalogue has corrected three times.
Reopened: this close was not a decision, it was a parser.
The issue was closed at 17:07:57 by the merge of #1873, 83 seconds after the comment above said "the issue stays open". Nothing was built and the field is still unpublished. PR #1873's body carries a heading describing the unblock condition — what would close #N — and Forgejo reads a close keyword followed by an issue number anywhere in a pull request body, with no regard for the sentence around it. The heading describing the unblock condition was the close instruction, aimed at this issue.
The same shape took the sibling request #743 at 16:06:14, from a body sentence that disclaimed the close. Both are back open.
What changed so it cannot recur silently:
scripts/check-closed-issues.jsnow reports any close that contradictsresearch/open-device-requests.json, which lists this issue as blocked on a vendor value, and it reads the closed list most-recently-updated first so a close on an old-numbered request is actually inside the audit window (the default order is by number, and #742 sat about 1,100 rows below it). The CI jobclosed-audit.ymlruns that on every push to main.Status unchanged: blocked on the interface-module slots per controller and the ports on each module type, per the comment above. The issue stays accepted, not answered.
Re-verified cheaply today, and this pass produces the negative the thread has been missing: Huawei publishes this exact field per model for the X series and does not publish it for the E series.
1. The datum that settles it
The document that gives #743 its port census - Huawei OceanProtect X Series Backup Appliance Technical White Paper, Issue 01 (2026-09-29), material page
https://e.huawei.com/en/documents/products/storage/eded1b265ecc4dfba2fbf8669b0b8213- states, in Chapter 3, a management-port count per model: X3000 ("four onboard GE ports, two onboard 10GE ports, two GE management/maintenance ports, and one serial port"), X6000 / X8000 / X8100 ("Each controller has two GE management/maintenance ports and one serial port"), X9000 / X9100 ("two management modules ... each management module has three GE ports, one USB port, and one serial port").The same file, dumped to 6,458 lines of text, contains the string
E8000zero times, andE6000zero times. So the omission is not an access failure and not an unsearched page: the vendor publishes this field for six OceanProtect models in one document and left the E series out of it.2. The E8000's own sources, re-read today
https://e.huawei.com/en/documents/products/storage/597eb7d1363f4b5da0d4ad663da1f7b6) - all four slides' text extracted this time, not only the specification slide. The only port rows remainNumber of Controllers per Node | 2andFront-End Port Types | 16/32/64 Gbit/s Fibre Channel; 10/25/40/100GE. No management row, no port count, no slot count. One detail worth keeping: its electrical table prices two enclosures - "NVMe controller enclosure: 86.1 mm x 447 mm x 920 mm" and "Controller enclosure: 86.1 mm x 447 mm x 820 mm" - and publishes a port list for neither.https://e.huawei.com/en/products/storage/oceanprotect/oceanprotect-e8000(200, 343,723 B): port types and protocols only, and its single document link is the datasheet above.EDOC1100486186, Hardware Installation GuideEDOC1100486185, and the Quick Installation GuideEDOC1100486182(id newly recorded this pass; all three listed for the model onhttps://support.huawei.com/enterprise/en/data-protection/oceanprotect-e8000-pid-261860234) - answer HTTP 302 with headerX-ALB-HIT: req-hd-bl, onsupport.huawei.com,info.support.huawei.comandsupport.huawei.cnalike.3. The one route to an E8000 census, recorded and deliberately not taken
The E8000-applicable OceanProtect Backup Storage 1.x and V200R001 Parts Replacement (
EDOC1100305200) groups the hardware out loud: its CRU/FRU table reads "Fan module (2 U controller enclosure and smart disk enclosure of OceanProtect X6000/X8000/E8000)", and it has a section titled "Replacing a Controller (OceanProtect X6000/X8000/E8000)" that says "The procedures for replacing controllers of this model are the same". If the E8000's 2 U controller enclosure is the X6000/X8000 one, then X8000's six slots per controller and 4-port 10GE electrical module would convert it in one step.Not taken, for two reasons: it is a family transfer (the precise error recorded in
research/oceanprotect-x-series-port-census.mdsections 4-5 and corrected in this catalogue repeatedly), andEDOC1100305200is itself ALB-bounced, so even the grouping above is a search-engine snippet rather than a page I fetched. Written down here so the next reader does not have to re-derive the temptation.4. Exact missing fields, unchanged
Per E8000 node: the controller-enclosure interface-module slot count, which slots accept front-end modules, and the ports-per-module count for the 10/25/40/100GE and 16/32/64 Gbit/s FC modules Huawei sells for it. The node has 2 controllers and its front-end port types are published; the counts are not.
What unblocks it: any one live section of
EDOC1100486186,EDOC1100486185orEDOC1100486182(one 8-hex slug opens the whole book, exactly asce945100did for X6000/X8000), or a stated configuration of the shape "E8000 node, 2 controllers, 4-port 10GE electrical module in IOM 0 and IOM 1", or Huawei's Spare Parts Query output for the model.Recording the pass in
research/oceanprotect-x-series-port-census.mdsections 14-15 (PR #1902). The X8100/X9100 half of that pair of requests is now modelled; this one stays open on the field above.Reopened: closed by a dry run that was not dry.
Both #742 and #743 were closed at 21:54:24-25 UTC by the
scripts/close-merged-issues.jsunder construction during the PR #1906 work — an early parser that matched the negated sentences in this thread ("Refs #742 ... it does not close the issue"). Its own report attributed the drop to PR #1905's housekeeping; #1905 merged four minutes earlier and closed nothing, so that attribution was wrong. The merged script requires a closing keyword to open a line or a sentence and to be un-negated, and it proposes neither issue today.Nothing was built for #742, and #743's spec is still open in PR #1902 — so the only correct state for both is open.
Reopened: the close carries no evidence, and the catalogue does not have this device.
Closed 2026-11-17 21:54:25 UTC, about fourteen minutes after the comment above re-affirmed the blocker, and with no comment of its own - the last comment here is still the 21:40 one.
Checked against
origin/mainat this run's merge base:git cat-file -e origin/main:devices/huawei/oceanprotect-e8000.json-> missing. The catalogue holdsdevices/huawei/oceanprotect-x8000.json, which is the X-series appliance from #741, and no E8000 spec at all.research/open-device-requests.jsonstill lists this issue as an open blocked request, and the table generated from it indocs/DEVICE_REGISTRY.mdprints it as open - so the repository's own record contradicted the close.A close that names neither a spec nor a reason is the shape
scripts/check-closed-issues.jsexists to catch; it could not see this one, because its closed-list read returned a single issue for the 60-day window. PR #1901, still open, is the fix for that read.Status unchanged: blocked on the E8000 front-end port count - interface-module slots per controller, and ports per module type. Unblocking evidence unchanged: any one rendered section of
EDOC1100486186,EDOC1100485347orEDOC1100486185, or a stated configuration such as "E8000 node, 2 controllers, 2 x 4-port 25GE IOM per controller".Re-checked 2026-11-18. Still blocked on the front-end census - but this pass counted the E8000's public document set instead of sampling it, and measured the one transfer that would otherwise look tempting. PR #1924 records it; it closes nothing.
1. The vendor file host carries exactly two E8000 documents, and both are data sheets. I took
https://e.huawei.com/robots.txt->https://e.huawei.com/sitemap-ebg-x-default.xml(HTTP 200, 1,803,417 B, 11,855<loc>), filtered the 5,522/documents/...entries to 224 under/en/documents/products/storage/, and titled all 224 (one request each, HTTP 200). Two are E8000:https://e.huawei.com/en/documents/products/storage/597eb7d1363f4b5da0d4ad663da1f7b6https://e.huawei.com/en/documents/products/storage/b2c7e3c34bcb4798812106bdd494fe91There is no public Product Description, Installation Guide or Quick Installation Guide. (The 26H2 sheet is new to this thread - earlier passes had only the C20 pptx.)
2. Both were read in full, not sampled. Downloaded anonymously from
e-file.huawei.com/marketingcloud/pep/asset/20000001/Material/..., extracted withpdftotext -layoutand the pptx<a:t>runs. Both containinterface module0,slot0,IOM0. The deepest row either prints isNumber of Controllers per Node 2besideFront-End Port Types 16/32/64 Gbit/s Fibre Channel; 10/25/40/100GE. A data sheet that never says slot cannot supply a slot count.3. hedex is login-gated for all three books that would answer.
https://support.huawei.com/hedex/api/navi/EDOC1100368651/first-level-tree(the family's "Downloadable Product Documentation Package", the only such link on the family page),.../EDOC1100486185/first-level-treeand.../EDOC1100486186/first-level-treeeach answer 401{"status":401,"message":"NOT LOGIN!"}; thetopic/navi-treevariant answers 400error!. The bare-document URLs in the form the product page itself emits (/enterprise/en/doc/EDOC1100486185?idPath=7919749|251366268|261607408|251366267|262802350) both answer 302 ->/view/notFound.4. The book whose title names E8000 is retired. OceanProtect Backup Storage 1.x and V200R001 Parts Replacement (
EDOC1100305200) is the one vendor book that pairs E8000 with X6000/X8000. Its two sections a search engine still returns now answer 302 ->/view/notFound:.../EDOC1100305200/6cc495b9/replacing-a-controller-oceanprotect-x6000-x8000-e8000and.../EDOC1100305200/50297f09/how-to-distinguish-crus-and-frus. The "one indexed section is enough" route that opened the X3000 book has nothing to open here.5. The trap, now measured rather than assumed. The readable
https://support.huawei.com/enterprise/en/doc/EDOC1100318730/ce945100/cabinet-layout-and-connection-planning(X6000/X8000 1.3.0 Installation Guide, 121,711 B) says "Each of controllers A and B provides six interface module slots" - and the stringE8000occurs zero times in its 473,037 characters of rendered text. Those six slots are the X6000/X8000 enclosure's; importing them into an E8000 spec would be the family-sibling error this catalogue has corrected before.6. Also closed: CC Huawei OceanProtect Software 1.6.0 Security Target (
https://www.commoncriteriaportal.org/files/epfiles/st_huawei_oceanprotect160_v1.10.pdf, 951,881 B) names E8000 in the evaluated set and carries no hardware port census.Unblocking evidence is unchanged: one rendered section of
EDOC1100486186/EDOC1100486185/EDOC1100485347pasted by a reader with a support session, or a stated configuration such as "E8000 node, 2 controllers, 2 x 4-port 25GE IOM per controller". Recorded inresearch/oceanprotect-e8000-public-document-set.mdand in the #742 row ofresearch/open-device-requests.json.Answered - the census is published, in Chinese. PR #1930 models both forms; it closes this issue on merge.
Every earlier pass searched the English document tree. The Chinese product page carries a different resource list, and its "白皮书" entry is the hardware book the hedex tree keeps behind a login, served anonymously from
e-file.huawei.com.Route (all 200, crawler user-agent):
单节点控制器数量 2,前端通道端口类型 16/32/64 Gb FC, 10/25/40/100 GbEattachmentUrlfield prints the asset pathapplication/pdf, 19,224,057 B, 141 pp, document version 03 (2026-05-06)The two fields this issue asked for. §3.1.3 OceanProtect E8000 (p.14): a 2 U two-controller 盘控一体 frame, 每控制框支持最多12个可热插拔的接口模块 - six per controller - and 每控有2个GE管理/维护口,1个串口. The per-model table (pp.7-9), row
接口卡, gives the ports on each module type: front-end 4-port 8G/16G/32G/64G FC, 4-port 10GE/25GE ETH, 2-port 40GE/100GE ETH; Scale-Out 4-port 25Gb RDMA; back-end 4-port 12Gb SAS (HDD form) or 100Gb RDMA, 2 ports per module (NVMe form). Slots x ports per module = the front-end census.One caveat, recorded in the specs rather than smoothed over: §3.1.3's first sentence opens with the X8000's model name - a copy-paste slip in the vendor text. The heading, figures 3-10/3-11 (E8000 NVMe 36-slot) and the 36-slot description are the E8000's, and the 12-module sentence is the one they introduce; it is the only vendor statement of the count, so both specs quote it verbatim and name the slip.
research/oceanprotect-e8000-cn-whitepaper.mdholds the full reading;devices/huawei/oceanprotect-e8000-hdd.jsonand...-nvme.jsonhold the specs (stated configuration: one 4-port 10GE/25GE module in IOM 0 and one in IOM 1 per controller, 16 front-end SFP28 ports, plus the 4 x 1GbE management ports). The next pass should start from a different issue - #130, #803, #807, #808 and #810 remain, and #742 leavesresearch/open-device-requests.json.